主题
语言

媒体模型目录

所有公开的 GetVibe API 媒体模型:图像、视频、TTS、音乐和 3D。卡片来自 GET /api/media-catalog:kind、variants、params、axes。POST /api/media/generate 的 curl 示例。

GetVibe 公共媒体模型的完整列表:图像、视频、配音(TTS)、音乐和 3D。下面的卡片会根据为 GET /api/media-catalog 提供支持的同一目录自动生成 — 向代码中添加模型后,文档和 storefront 会同步更新。

定价的权威来源: GET /api/media-catalog 价格表会区分作业总价和单位费率:在 priceScope = unit_rate 的行中,`price` 为 null,单位费率位于 `unitPrice`。对于 jobTotalStatus = metered 或 input_dependent 的情况,请使用 POST /api/media/price 报价,并提供实际输入。失败的作业不收费;GET /api/media/jobs/{id} 会返回最终金额。

如何读取 /api/media-catalog 响应

单个模型的字段
参数类型描述
modelfieldPOST /api/media/generate 的 model 字段使用的公共 id。
kindfieldimage | video | audio | music | 3d.
unitfield计费单位:按图像、按秒、按片段等计费。
notefield功能简介。
inputTypesfield允许的输入类型:文本、图像、视频、音频。
paramsfield旧版:opts 参数(key、type、default、options)。
fromRubfield各变体/参数中最低的“from”价格;当 priceStatus = exact 时为精确价格。
priceStatusfieldexact — 报价等于费用;estimate — 提交后根据计量结果计费;unavailable — 没有价格。
estimateReasonfield估算依据:据此计算费用的计量值。
priceTableUrlfield模型完整价格表链接:GET /api/media/price-table。
tiersfield查看目录实时 JSON。
variantsfield带有 priceRub 的 addressAs 路由(例如:base:pixel、base:flow)。
axesfield供 UI 和计算器使用的统一定价维度。
inputSchemafield用于客户端验证的 prompt + opts 字段结构。
priceUnavailablefieldtrue — 该路由无法调用(生成时返回 503)。
pricesAreEstimatefield当模型的至少一个变体为估算值时为 true。
pricingKindfieldfixed | gpu_runtime etc. — quote type.
estimatedPriceRubfield查看目录实时 JSON。
priceIsEstimatefield查看目录实时 JSON。
priceWarningfield查看目录实时 JSON。
generationAvailablefield对于 displayOnly 和 gpu_runtime,在未结算时为 false。
unavailableReasonfield查看目录实时 JSON。
displayOnlyfield不支持调度的参考模型(无法生成)。
curl "https://getvibeapi.com/api/media-catalog"
# public, no authorization

经销商价格表

GET /api/media/price-table?model=<id>(公开)会返回 opts、price、unitPrice、priceScope、jobTotalStatus 和 requiredInputs。`price` 是整个作业的金额;在 priceScope = unit_rate 的行中,它为 null,单位费率及其 `unit` 位于 `unitPrice` 中。requiredInputs 会列出额外的请求字段:allOf、anyOf 和 exactlyOneOf;prompt 位于顶层,其他字段位于 opts 中。requiresInput 会指出同一验证流程中缺少的第一个字段,该验证由 POST /api/media/price 执行:prompt 或输入文件。additionalFormsPossible 表示此列表并非完整的验证架构。请使用 POST /api/media/price,根据实际输入报价。

竞争对手处经常搜索的模型系列

Seedance、Wan、Hailuo、Suno、TTS 和 3D 已在 GetVibe 目录中:

  • Seedance(ByteDance): seedance-2.0seedance-1seedance-2-0-mini
  • Wan(Alibaba): wan-2-7wan-2-6wan-2-2wan-animatewan-speech-to-videowan-2-7-image
  • Hailuo(MiniMax): hailuo-2-3
  • Suno(音乐): suno
  • TTS(ElevenLabs): tts
  • 3D(Hunyuan、Tripo、Meshy、Rodin、Trellis): hunyuan3dhunyuan3d-2hunyuan3d-2mvtripomeshyrodintrellis

模型卡片 (204)

按媒体类型划分的汇总表,以及包含参数和示例 curl 的可展开卡片。带有 :pixel、:flow 或 :rep 后缀的变体会原样传入 model 字段(参见 variants[].addressAs)。

Images (54)

model输入单位变体opts 参数
seedream-5-0-protext每张图片seedream-5-0-pro:whiteresolution
grok-imagine-2.0-exttext, image每张图片grok-imagine-2.0-ext:white (unavailable)mode, image_url, resolution
grok-imagine-image-2.0text, image每张图片grok-imagine-image-2.0:whiteresolution, quality, image_url
grok-imagine-image-qualitytext, image每张图片grok-imagine-image-quality:whiteaspect_ratio, resolution, image_url, input_urls
midjourneytext, image每张图片midjourney:white (unavailable)operation, image_url
wan2.7-image-protext, image每张图片wan2.7-image-pro:whiteresolution, image_url, input_urls
seedream-5-0-flashtext, image每张图片seedream-5-0-flash:white, seedream-5-0-flash:flowaspect_ratio, resolution, output_format, image_url
flux-2-maxtext每张图片flux-2-max:white, flux-2-max:flowaspect_ratio, resolution, output_format, safety_tolerance, prompt_upsampling, seed
recraft-v4-1text每张图片recraft-v4-1tier, aspect_ratio
ideogram-v4text每张图片ideogram-v4rendering_speed, aspect_ratio
ideogram-4-5text, image每张图片ideogram-4-5quality, size, image_url, input_urls
ideogram-4-5-precise-edittext, image每张图片ideogram-4-5-precise-editquality, image_url, input_urls
krea-3text, image每张图片krea-3variant, resolution, aspect_ratio, reasoning_level, output_format, image_url, input_urls, negative_prompt, seed
flux-3-kleintext, image每张图片flux-3-klein, flux-3-klein:flowresolution, aspect_ratio, output_format, enable_prompt_expansion, image_url, input_urls, seed
muse-imagetext, image每张图片muse-imageaspect_ratio, output_format, image_url, input_urls
seedream-5-5text, image每张图片seedream-5-5resolution, aspect_ratio, output_format, background, image_url, input_urls
gpt-image-2text, image每张图片gpt-image-2:white, gpt-image-2, gpt-image-2:yellow (unavailable)aspect_ratio, resolution, mode, image_url, input_urls
gpt-image-2-5text, image每张图片gpt-image-2-5:white, gpt-image-2-5, gpt-image-2-5:yellow (unavailable)aspect_ratio, resolution, tier, mode, image_url, input_urls
gpt-image-2-6text, image每张图片gpt-image-2-6:white, gpt-image-2-6:canvasaspect_ratio, resolution, quality, complexity, tier, mode, image_url, input_urls
gpt-image-1-5text, image每张图片—aspect_ratio, mode, quality, image_url, input_urls
nano-banana-protext, image每张图片nano-banana-pro:white, nano-banana-pro, nano-banana-pro:yellow (unavailable)aspect_ratio, resolution, mode, image_url, input_urls
nano-bananatext, image每张图片nano-banana:白色(不可用)、nano-bananaaspect_ratio, mode, image_url, input_urls
imagen4text每张图片—aspect_ratio, tier, negative_prompt, seed
nano-banana-2.1text, image每张图片nano-banana-2.1:white, nano-banana-2.1, nano-banana-2.1:yellow (unavailable)aspect_ratio, resolution, output_format, image_url, input_urls
nano-banana-2text, image每张图片nano-banana-2:white (unavailable), nano-banana-2, nano-banana-2:yellow (unavailable)aspect_ratio, tier, resolution, image_url, input_urls
grok-imaginetext, image每张图片grok-imagine:white, grok-imagineaspect_ratio, mode, image_url, enable_pro
ideogram-v3text, image每张图片ideogram-v3:pixel, ideogram-v3:repaspect_ratio, mode, rendering_speed, image_url, style, expand_prompt, seed, negative_prompt, strength, num_images
ideogram-charactertext, image每张图片ideogram-character, ideogram-character:repaspect_ratio, mode, image_url, input_urls, rendering_speed, style, expand_prompt, seed, negative_prompt, strength, num_images
flux-kontexttext, image每张图片—aspect_ratio, tier, image_url
flux-2text, image每张图片flux-2, flux-2:flow (unavailable)aspect_ratio, tier, mode, resolution, image_url, input_urls
flux-3-imagetext, image每张图片flux-3-image:whiteaspect_ratio, resolution, image_url, input_urls
krea-2text, image每张图片krea-2tier, style_reference, aspect_ratio, creativity, image_url, seed
flux-1.1-protext每张图片flux-1.1-pro, flux-1.1-pro:amber (unavailable)aspect_ratio, seed
flux-filltext, image每张图片flux-filltier, image_url, mask_url, seed
seedreamtext, image每张图片—aspect_ratio, version, mode, image_url, input_urls, guidance, image_resolution, max_images, seed
seedream-5text, image每张图片—aspect_ratio, tier, mode, quality, image_url, input_urls
seedream-layerimage每张图片—image_url, size, output_format, prompt
seedream-4-5text, image每张图片seedream-4-5, seedream-4-5:amber (unavailable)aspect_ratio, mode, quality, image_url, input_urls
qwen-imagetext, image每张图片—aspect_ratio, version, mode, image_url, negative_prompt, strength, guidance, seed, num_inference_steps, acceleration
qwen3-imagetext, image每张图片qwen3-image:white, qwen3-image, qwen3-image:amber (unavailable)aspect_ratio, resolution, mode, image_url, input_urls, negative_prompt, seed
qwen3-protext, image每张图片qwen3-pro, qwen3-pro:amber (unavailable)aspect_ratio, resolution, mode, image_url, input_urls, negative_prompt, seed
qwen-image-2-1text, image每张图片qwen-image-2-1aspect_ratio, resolution, mode, image_url, input_urls, mask_url, background, output_format, enhance_prompt, seed
recraftimage每张图片—mode, image_url
image-upscaleimage每张图片—upscale_factor, image_url
z-imagetext每张图片—aspect_ratio
z-image-turbotext每张图片z-image-turbo:whiteaspect_ratio, resolution, prompt_extend
hunyuan-image-3-5text, image每张图片hunyuan-image-3-5:whiteaspect_ratio, resolution, image_url, input_urls
wan-2-7-imagetext, image每张图片wan-2-7-image, wan-2-7-image:white (unavailable)aspect_ratio, tier, mode, enable_sequential, resolution, n, seed, watermark, thinking_mode, color_palette, bbox_list, image_url, input_urls
grok-imagine-image-2-0text, image每张图片—mode, aspect_ratio, image_urls, image_url
minimax-h3-imagetext, image每张图片—mode, aspect_ratio, resolution, width, height, expand_prompt, output_format, seed, image_url, image_urls
topaz-upscale-generativeimage每张图片topaz-upscale-generativeimage_url, model, source_megapixels, output_megapixels, upscale_factor
topaz-sharpenimage每张图片topaz-sharpenimage_url, model, source_megapixels
topaz-denoiseimage每张图片topaz-denoiseimage_url, model, source_megapixels
topaz-upscale-precisionimage每张图片topaz-upscale-precisionimage_url, model, source_megapixels, output_megapixels, upscale_factor, output_format, face_enhancement
seedream-5-0-pro

Seedream 5.0 Pro(ByteDance)— 支持 1K 或 2K 文生图。按分辨率逐张计费。

变体(addressAs)
参数类型描述
seedream-5-0-pro:whitewhiteWhite · Available
opts 参数
参数类型描述
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
  • variants: seedream-5-0-pro:white
  • resolution: 1K | 2K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedream-5-0-pro","prompt":"A beautiful landscape in a digital illustration style.","opts":{"resolution":"1K"}}'
grok-imagine-2.0-ext

Grok Imagine 2.0 区域编辑 — 暂时不可用。

变体(addressAs)
参数类型描述
grok-imagine-2.0-ext:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
modeselect模式 · 默认值:region_edit · 选项:region_edit
image_urlreq.text图像(URL)
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
  • variants: grok-imagine-2.0-ext:white
  • mode: region_edit
  • resolution: 1K | 2K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-imagine-2.0-ext","prompt":"A beautiful landscape in a digital illustration style.","opts":{"mode":"region_edit","image_url":"https://example.com/input.jpg","resolution":"1K"}}'
grok-imagine-image-2.0

Grok Imagine Image 2.0 (xAI) — 支持 1K、基础质量的文生图,或根据一张源图像进行编辑。按图像计费;源图像会产生附加费用,该费用已包含在价格报价中。

变体(addressAs)
参数类型描述
grok-imagine-image-2.0:whitewhiteWhite · Available
opts 参数
参数类型描述
resolutionselect分辨率 · 默认值:1K · 选项:1K
qualityselect质量 · 默认值:low · 选项:low
image_urltext源图(URL,每张输入图收取附加费)
  • variants: grok-imagine-image-2.0:white
  • resolution: 1K
  • 质量:低

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-imagine-image-2.0","prompt":"A beautiful landscape in a digital illustration style.","opts":{"resolution":"1K","quality":"low"}}'
grok-imagine-image-quality

Grok Imagine Image Quality(xAI)——在 1K 或 2K 下进行文生图和参考图编辑(最多 3 张图)。按图计价;每张源图都会产生附加费,该费用已包含在报价中。

变体(addressAs)
参数类型描述
grok-imagine-image-quality:whitewhiteWhite · Available
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: auto · options: auto, 1:1, 3:4, 4:3, 9:16, 16:9, 2:3, 3:2, 9:19.5, 19.5:9, 9:20, 20:9, 1:2, 2:1
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
image_urltext源图(URL,每张输入图收取附加费)
input_urlsarray源图(URL,最多 3 张,适用附加费)
  • variants: grok-imagine-image-quality:white
  • aspect_ratio: auto | 1:1 | 3:4 | 4:3 | 9:16 | 16:9 | 2:3 | 3:2 | 9:19.5 | 19.5:9 | 9:20 | 20:9 | 1:2 | 2:1
  • resolution: 1K | 2K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-imagine-image-quality","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"auto","resolution":"1K","input_urls":["https://example.com/input.jpg"]}}'
midjourney

Midjourney:暂时不可用。

变体(addressAs)
参数类型描述
midjourney:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
operationselect操作 · 默认值:imagine · 选项:imagine、action、turbo
image_urltext图像(URL)
  • variants: midjourney:white
  • operation: imagine | action | turbo

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"midjourney","prompt":"A beautiful landscape in a digital illustration style.","opts":{"operation":"imagine"}}'
wan2.7-image-pro

Wan 2.7 Image Pro(Alibaba):支持最多 9 个参考图像的图像生成和编辑。按图像计费。

变体(addressAs)
参数类型描述
wan2.7-image-pro:whitewhiteWhite · Available
opts 参数
参数类型描述
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
image_urltext图像(URL,用于 i2i)
input_urlsarray图像(1–9 个 URL,用于 i2i)
  • variants: wan2.7-image-pro:white
  • resolution: 1K | 2K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan2.7-image-pro","prompt":"A beautiful landscape in a digital illustration style.","opts":{"resolution":"1K","input_urls":["https://example.com/input.jpg"]}}'
seedream-5-0-flash

Seedream 5.0 Flash(ByteDance)— 快速文生图和单参考图编辑。输出支持 1K / 1.5K / 2K,10 种宽高比或自动设置。所有分辨率下每张图片均为统一价格。

变体(addressAs)
参数类型描述
seedream-5-0-flash:whitewhiteWhite · Available
seedream-5-0-flash:flowflowFlow · 可用
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: auto · options: auto, 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, 2:1, 1:2, 21:9
resolutionselect分辨率 · 默认值:2K · 选项:1K、1.5K、2K
output_formatselect格式 · 默认值:jpeg · 选项:jpeg、png
image_urltext图像(URL,用于编辑)
  • variants: seedream-5-0-flash:white, seedream-5-0-flash:flow
  • aspect_ratio: auto | 1:1 | 4:3 | 3:4 | 16:9 | 9:16 | 3:2 | 2:3 | 2:1 | 1:2 | 21:9
  • resolution: 1K | 1.5K | 2K
  • output_format: jpeg | png

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedream-5-0-flash","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"auto","resolution":"2K","output_format":"jpeg"}}'
flux-2-max

FLUX.2 [max](Black Forest Labs)— FLUX.2 中面向最高细节表现的顶级层级,支持文生图。按输出尺寸计费:1MP / 2MP / 3MP / 4MP。

变体(addressAs)
参数类型描述
flux-2-max:whitewhiteWhite · Available
flux-2-max:flowflowFlow · 可用
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, 21:9, 9:21
resolutionselectResolution · default: 1MP · options: 1MP, 2MP, 3MP, 4MP
output_formatselect格式 · 默认值:jpeg · 选项:jpeg、png、webp
safety_toleranceselectSafety tolerance (0 strictest, 5 most permissive) · default: 2 · options: 0, 1, 2, 3, 4, 5
prompt_upsamplingbool提示词上采样 · 默认值:false
seednumberSeed(可选)
  • variants: flux-2-max:white, flux-2-max:flow
  • aspect_ratio: 1:1 | 4:3 | 3:4 | 16:9 | 9:16 | 3:2 | 2:3 | 21:9 | 9:21
  • resolution: 1MP | 2MP | 3MP | 4MP
  • output_format: jpeg | png | webp
  • safety_tolerance: 0 | 1 | 2 | 3 | 4 | 5

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-2-max","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1MP","output_format":"jpeg","safety_tolerance":"2"}}'
recraft-v4-1

Recraft V4.1 — 用于设计、插画、照片级真实感和清晰排版的文生图模型。提供三个层级:Flash(最便宜且最快)、Standard 和 Pro(细节最多)。按层级以每张图片计费。

变体(addressAs)
参数类型描述
recraft-v4-1flowFlow · 可用
opts 参数
参数类型描述
tierselect层级 · 默认值:standard · 选项:flash, standard, pro
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 4:3, 3:4, 16:9, 9:16
  • variants: recraft-v4-1
  • tier: flash | standard | pro
  • aspect_ratio: 1:1 | 4:3 | 3:4 | 16:9 | 9:16

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"recraft-v4-1","prompt":"A beautiful landscape in a digital illustration style.","opts":{"tier":"standard","aspect_ratio":"1:1"}}'
ideogram-v4

Ideogram 4.0 — Ideogram 于 2026 年 6 月推出的新一代生成模型:支持精确排版和图像内文本、海报、徽标及插画;输出约为 2K。提供三种速度:Turbo、Balanced、Quality。按速度和图像数量计费。

变体(addressAs)
参数类型描述
ideogram-v4canvasCanvas · 可用
opts 参数
参数类型描述
rendering_speedselect速度 · 默认值:balanced · 可选值:turbo、balanced、quality
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 16:10, 10:16, 2:1, 1:2
  • variants: ideogram-v4
  • rendering_speed: turbo | balanced | quality
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3 | 16:10 | 10:16 | 2:1 | 1:2

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"ideogram-v4","prompt":"A beautiful landscape in a digital illustration style.","opts":{"rendering_speed":"balanced","aspect_ratio":"1:1"}}'
ideogram-4-5

Ideogram 4.5 — 最新一代 Ideogram:支持精准排版和图像内文字、海报、徽标及源图像编辑(最多 5 张参考图像)。提供四个质量等级:极低、低、中、高。按质量等级逐张图像计费;源图像编辑按单独的费率计费。

变体(addressAs)
参数类型描述
ideogram-4-5canvasCanvas · 可用
opts 参数
参数类型描述
qualityselect质量 · 默认值:medium · 可选值:very_low, low, medium, high
sizeselectSize · default: 1024x1024 · options: 1024x1024, 1344x768, 768x1344, 1280x896, 896x1280, 1536x640, 640x1536
image_urltext源图像(URL,可选)
input_urlsarray源图像(URL,最多 5 张)
  • variants: ideogram-4-5
  • quality: very_low | low | medium | high
  • size: 1024x1024 | 1344x768 | 768x1344 | 1280x896 | 896x1280 | 1536x640 | 640x1536

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"ideogram-4-5","prompt":"A beautiful landscape in a digital illustration style.","opts":{"quality":"medium","size":"1024x1024","input_urls":["https://example.com/input.jpg"]}}'
ideogram-4-5-precise-edit

Ideogram 4.5 Precise Edit — 根据文本指令忠实编辑源图像,同时保留原有构图。必须提供源图像(URL)。提供四个质量等级:极低、低、中、高。按质量等级逐张图像计费。

变体(addressAs)
参数类型描述
ideogram-4-5-precise-editcanvasCanvas · 可用
opts 参数
参数类型描述
qualityselect质量 · 默认值:medium · 可选值:very_low, low, medium, high
image_urlreq.text源图像(URL)
input_urlsarray源图像(URL,最多 5 张)
  • variants: ideogram-4-5-precise-edit
  • quality: very_low | low | medium | high

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"ideogram-4-5-precise-edit","prompt":"A beautiful landscape in a digital illustration style.","opts":{"quality":"medium","image_url":"https://example.com/input.jpg","input_urls":["https://example.com/input.jpg"]}}'
krea-3

Krea 3 — Krea 最新一代模型(2026 年 10 月):支持文本生成图像,以及根据参考图像(最多 8 张)使用自然语言进行编辑。提供两种变体:Quality 和 Speed。支持 1K(约 1 MP)或 2K(约 4 MP)。按尺寸逐张计费;每张参考图像会按相同尺寸额外增加一张图像的价格。

变体(addressAs)
参数类型描述
krea-3flowFlow · 可用
opts 参数
参数类型描述
variantselect选项 · 默认值:quality · 选项:quality、speed
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3
reasoning_levelselect提示词推理 · 默认值:low · 选项:low、medium
output_formatselect格式 · 默认值:png · 选项:png、jpeg
image_urltext参考图像(用于编辑的 URL)
input_urlsarray参考图像(URL,最多 8 张)
negative_prompttext负面提示词(仅限 Quality)
seednumberSeed(可选)
  • variants: krea-3
  • variant: quality | speed
  • resolution: 1K | 2K
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3
  • reasoning_level: low | medium
  • output_format: png | jpeg

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"krea-3","prompt":"A beautiful landscape in a digital illustration style.","opts":{"variant":"quality","resolution":"1K","aspect_ratio":"1:1","reasoning_level":"low","output_format":"png","input_urls":["https://example.com/input.jpg"]}}'
flux-3-klein

FLUX 3 Klein — Black Forest Labs 的 FLUX 3 系列中价格最实惠的模型(2026 年 10 月):支持文生图,以及根据参考图进行编辑(最多 10 张)。支持 1K(约 1 MP)、2K(约 4 MP)或 4K(最高 16 MP)。根据输出尺寸按图计费;参考图不会改变价格。

变体(addressAs)
参数类型描述
flux-3-kleinpixelPixel · 可用
flux-3-klein:flowflowFlow · 可用
opts 参数
参数类型描述
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K、4K
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9, 9:21, auto
output_formatselect格式 · 默认值:png · 选项:png、jpeg、webp
enable_prompt_expansionbool自动扩展提示词 · 默认值:true
image_urltext参考图像(用于编辑的 URL)
input_urlsarray参考图(URL,最多 10 张)
seednumberSeed(可选)
  • variants: flux-3-klein, flux-3-klein:flow
  • resolution: 1K | 2K | 4K
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3 | 21:9 | 9:21 | auto
  • output_format: png | jpeg | webp

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-3-klein","prompt":"A beautiful landscape in a digital illustration style.","opts":{"resolution":"1K","aspect_ratio":"1:1","output_format":"png","input_urls":["https://example.com/input.jpg"]}}'
muse-image

Muse Image by Meta — 支持文生图,以及使用自然语言根据参考图进行编辑(最多 10 张)。模型会自行选择尺寸(约 2.5 MP,例如 1:1 时为 1600×1600)。无论宽高比如何,每张图价格相同。

变体(addressAs)
参数类型描述
muse-imageflowFlow · 可用
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: auto · options: auto, 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9
output_formatselect格式 · 默认值:png · 选项:png、jpeg、webp
image_urltext参考图像(用于编辑的 URL)
input_urlsarray参考图(URL,最多 10 张)
  • variants: muse-image
  • aspect_ratio: auto | 1:1 | 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3 | 21:9
  • output_format: png | jpeg | webp

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"muse-image","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"auto","output_format":"png","input_urls":["https://example.com/input.jpg"]}}'
seedream-5-5

ByteDance 的 Seedream 5.5 — 支持文生图和基于参考图像的编辑(最多 10 张)。支持 1K / 1.5K / 2K 输出、5 种宽高比或自动选择、PNG 或 JPEG 格式,以及可选的透明背景。每张图像的价格取决于分辨率(2K 价格更高);每张参考图像均收取固定附加费。

变体(addressAs)
参数类型描述
seedream-5-5flowFlow · 可用
opts 参数
参数类型描述
resolutionselect分辨率 · 默认值:2K · 选项:1K、1.5K、2K
aspect_ratioselect宽高比 · 默认值:auto · 可选值:auto, 1:1, 4:3, 3:4, 16:9, 9:16
output_formatselect格式 · 默认值:png · 选项:png、jpeg
backgroundselect背景 · 默认值:auto · 可选值:auto, opaque, transparent
image_urltext参考图像(用于编辑的 URL)
input_urlsarray参考图(URL,最多 10 张)
  • variants: seedream-5-5
  • resolution: 1K | 1.5K | 2K
  • aspect_ratio: auto | 1:1 | 4:3 | 3:4 | 16:9 | 9:16
  • output_format: png | jpeg
  • background: auto | opaque | transparent

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedream-5-5","prompt":"A beautiful landscape in a digital illustration style.","opts":{"resolution":"2K","aspect_ratio":"auto","output_format":"png","background":"auto","input_urls":["https://example.com/input.jpg"]}}'
gpt-image-2

GPT Image 2,图像生成与编辑。支持文生图和图生图(编辑/修复)。

变体(addressAs)
参数类型描述
gpt-image-2:whitewhiteWhite · Available
gpt-image-2pixelPixel · 可用
gpt-image-2:yellowyellowYellow · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K、4K
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
image_urltext图像(URL,用于编辑模式)
input_urlsarray图像(1–16 个 URL,用于编辑模式)
  • variants: gpt-image-2:white, gpt-image-2, gpt-image-2:network, gpt-image-2:yellow
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3 | 21:9
  • resolution: 1K | 2K | 4K (affects price)
  • mode: text-to-image | image-to-image (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-image-2","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1K","mode":"text-to-image","input_urls":["https://example.com/input.jpg"]}}'
gpt-image-2-5

GPT Image 2.5,图像生成与编辑。支持文生图和图生图。提供两种 Pixel variant:Flare(快速,默认)和 Sunburst(高精度)。

变体(addressAs)
参数类型描述
gpt-image-2-5:whitewhiteWhite · Available
gpt-image-2-5pixelPixel · 可用
gpt-image-2-5:yellowyellowYellow · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K、4K
tierselect选项 · 默认值:flare · 选项:flare、sunburst
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
image_urltext图像(URL,用于编辑模式)
input_urlsarray图像(1–16 个 URL,用于编辑模式)
  • variants: gpt-image-2-5:white, gpt-image-2-5, gpt-image-2-5:network, gpt-image-2-5:yellow
  • tiers: 1K, 2K, 4K
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3 | 21:9
  • resolution: 1K | 2K | 4K (affects price)
  • tier: flare | sunburst (affects price)
  • mode: text-to-image | image-to-image (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-image-2-5","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1K","tier":"flare","mode":"text-to-image","input_urls":["https://example.com/input.jpg"]}}'
gpt-image-2-6

GPT Image 2.6 — 图像生成与编辑。支持文生图和图生图。提供 Flare(快速,默认)和 Sunburst(精准),支持复杂度和质量控制。主路由支持 1K、2K 和 4K,并以 Canvas 作为备用。

变体(addressAs)
参数类型描述
gpt-image-2-6:whitewhiteWhite · Available
gpt-image-2-6:canvascanvasCanvas · 可用
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K、4K
qualityselect质量 · 默认值:high · 可选值:low, medium, high
complexityselect复杂度 · 默认值:high · 可选值:low, medium, high
tierselect选项 · 默认值:flare · 选项:flare、sunburst
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
image_urltext图像(URL,用于编辑模式)
input_urlsarray图像(1–16 个 URL,用于编辑模式)
  • variants: gpt-image-2-6:white, gpt-image-2-6:canvas
  • 档位:1K · low, 1K · medium, 1K · high, 2K · low, 2K · medium, 2K · high, 4K · low, 4K · medium, 4K · high
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3 | 21:9
  • resolution: 1K | 2K | 4K (affects price)
  • quality: low | medium | high (affects price)
  • complexity: low | medium | high (affects price)
  • tier: flare | sunburst (affects price)
  • mode: text-to-image | image-to-image (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-image-2-6","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1K","quality":"high","complexity":"high","tier":"flare","mode":"text-to-image","input_urls":["https://example.com/input.jpg"]}}'
gpt-image-1-5

GPT Image 1.5,图像生成与编辑。

opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:1:1 · 选项:1:1、2:3、3:2
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
qualityselect质量 · 默认值:medium · 选项:medium、high
image_urltext图像(URL,用于编辑模式)
input_urlsarray图像(1–16 个 URL,用于编辑模式)
  • aspect_ratio: 1:1 | 2:3 | 3:2
  • mode: text-to-image | image-to-image (affects price)
  • quality: medium | high (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-image-1-5","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","mode":"text-to-image","quality":"medium","input_urls":["https://example.com/input.jpg"]}}'
nano-banana-pro

Nano Banana Pro —— Google 图像生成和编辑。支持文生图以及基于提示词的编辑(移除/替换元素)。

变体(addressAs)
参数类型描述
nano-banana-pro:whitewhiteWhite · Available
nano-banana-propixelPixel · 可用
nano-banana-pro:yellowyellowYellow · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K、4K
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、edit
image_urltext图像(URL,用于编辑模式)
input_urlsarray图像(1–8 个 URL,用于编辑模式)
  • variants: nano-banana-pro:white, nano-banana-pro, nano-banana-pro:yellow
  • tiers: 1K, 2K, 4K
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • resolution: 1K | 2K | 4K
  • mode: text-to-image | edit (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"nano-banana-pro","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1K","mode":"text-to-image","input_urls":["https://example.com/input.jpg"]}}'
nano-banana

Nano Banana(Gemini 2.5 Flash Image)——快速图像生成(Pixel google/nano-banana)和编辑(google/nano-banana-edit、image_urls)。不进行分层路由。

变体(addressAs)
参数类型描述
nano-banana:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
nano-bananapixelPixel · 可用
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、edit
image_urltext图像(URL,用于编辑模式)
input_urlsarray图像(1–10 个 URL,用于编辑模式)
  • 变体:nano-banana:white、nano-banana
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • mode: text-to-image | edit (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"nano-banana","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","mode":"text-to-image","input_urls":["https://example.com/input.jpg"]}}'
imagen4

Google Imagen 4 — 高质量。

opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
tierselect质量 · 默认值:standard · 选项:fast、standard、ultra
negative_prompttext负面提示词(可选)
seednumberSeed(可选)
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • tier: fast | standard | ultra (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"imagen4","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","tier":"standard"}}'
nano-banana-2.1

Nano Banana 2.1 —— Google 图像生成和编辑,支持最多 10 张参考图像、全景宽高比、1K/2K/4K 分辨率以及 PNG/JPG 输出。

变体(addressAs)
参数类型描述
nano-banana-2.1:whitewhiteWhite · Available
nano-banana-2.1pixelPixel · 可用
nano-banana-2.1:yellowyellowYellow · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: auto · options: 1:1, 2:3, 3:2, 1:4, 4:1, 3:4, 4:3, 4:5, 5:4, 1:8, 8:1, 9:16, 16:9, 21:9, auto
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K、4K
output_formatselect格式 · 默认值:jpg · 选项:png、jpg
image_urltext图像(URL,参考图)
input_urlsarray图像(1–10 个 URL,参考图)
  • variants: nano-banana-2.1:white, nano-banana-2.1, nano-banana-2.1:yellow
  • tiers: 1K, 2K, 4K
  • aspect_ratio: 1:1 | 2:3 | 3:2 | 1:4 | 4:1 | 3:4 | 4:3 | 4:5 | 5:4 | 1:8 | 8:1 | 9:16 | 16:9 | 21:9 | auto
  • resolution: 1K | 2K | 4K
  • output_format: png | jpg

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"nano-banana-2.1","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"auto","resolution":"1K","output_format":"jpg","input_urls":["https://example.com/input.jpg"]}}'
nano-banana-2

Nano Banana 2 —— Google 根据文本或参考图像生成图像。Lite → nano-banana-2-lite;Standard → nano-banana-2(上游支持 1K/2K/4K 分辨率)。

变体(addressAs)
参数类型描述
nano-banana-2:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
nano-banana-2pixelPixel · 可用
nano-banana-2:yellowyellowYellow · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
tierselect质量 · 默认值:standard · 选项:lite、standard
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K、4K
image_urltext图像(URL,参考图)
input_urlsarray图像(1–10 个 URL,参考图)
  • variants: nano-banana-2:white, nano-banana-2, nano-banana-2:network, nano-banana-2:yellow
  • 层级:Lite、Standard · 1K、Standard · 2K、Standard · 4K
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • tier: lite | standard
  • resolution: 1K | 2K | 4K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"nano-banana-2","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","tier":"standard","resolution":"1K","input_urls":["https://example.com/input.jpg"]}}'
grok-imagine

Grok Imagine — 由 xAI 提供的图像生成和图生图功能。每张图像 4 credits;I2I 需要输入图像。

变体(addressAs)
参数类型描述
grok-imagine:whitewhiteWhite · Available
grok-imaginepixelPixel · 可用
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
image_urltext图像(URL,用于 i2i)
enable_proboolPro(质量优先于速度)· 默认值:false
  • 变体:grok-imagine:white、grok-imagine
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • mode: text-to-image | image-to-image (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-imagine","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","mode":"text-to-image"}}'
ideogram-v3

Ideogram v3 — 出色的图像文字渲染效果。模式:generation、remix 和 reframe(后两种仅限 Pixel);不使用蒙版编辑 — 仅限 Canvas。

变体(addressAs)
参数类型描述
ideogram-v3:pixelpixelPixel · 可用
ideogram-v3:repcanvasCanvas · 可用
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、edit、remix(仅限 Pixel)、reframe(仅限 Pixel)
rendering_speedselect速度 · 默认值:DEFAULT · 可选值:TURBO、DEFAULT、QUALITY
image_urltext图像(URL,用于编辑/混剪/重构)
styleselect风格 · 默认值:AUTO · 可选值:AUTO、GENERAL、REALISTIC、DESIGN
expand_promptbool展开提示词 · 默认值:false
seednumberSeed
negative_prompttext负面提示词
strengthselectStrength (remix) · default: 0.8 · options: 0.8, 0.2, 0.4, 0.6, 1
num_imagesselect图像数量 · 默认值:1 · 选项:1、2、3、4
  • variants: ideogram-v3:pixel, ideogram-v3:rep
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • mode: text-to-image | edit | remix | reframe (affects price)
  • rendering_speed: TURBO | DEFAULT | QUALITY (affects price)
  • style: AUTO | GENERAL | REALISTIC | DESIGN
  • strength: 0.8 | 0.2 | 0.4 | 0.6 | 1
  • num_images: 1 | 2 | 3 | 4

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"ideogram-v3:pixel","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","mode":"text-to-image","rendering_speed":"DEFAULT","style":"AUTO","strength":"0.8","num_images":"1"}}'
ideogram-character

Ideogram Character — 风格化角色。不使用蒙版编辑角色 — 仅限 Canvas。

变体(addressAs)
参数类型描述
ideogram-characterpixelPixel · 可用
ideogram-character:repcanvasCanvas · 可用
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
modeselect模式 · 默认值:character · 选项:character、edit、remix
image_urltext图像 URL(角色参考图;用于编辑/混剪的画布)
input_urlsarray角色参考(URL,用于编辑/混合)
rendering_speedselect速度 · 默认值:DEFAULT · 可选值:TURBO、DEFAULT、QUALITY
styleselect风格 · 默认值:AUTO · 可选值:AUTO、REALISTIC、FICTION
expand_promptbool展开提示词 · 默认值:false
seednumberSeed
negative_prompttext负面提示词
strengthselectStrength (remix) · default: 0.8 · options: 0.8, 0.2, 0.4, 0.6, 1
num_imagesselect图像数量 · 默认值:1 · 选项:1、2、3、4
  • 变体:ideogram-character、ideogram-character:rep
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • mode: character | edit | remix (affects price)
  • rendering_speed: TURBO | DEFAULT | QUALITY
  • style: AUTO | REALISTIC | FICTION
  • strength: 0.8 | 0.2 | 0.4 | 0.6 | 1
  • num_images: 1 | 2 | 3 | 4

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"ideogram-character","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","mode":"character","input_urls":["https://example.com/input.jpg"],"rendering_speed":"DEFAULT","style":"AUTO","strength":"0.8","num_images":"1"}}'
flux-kontext

Flux Kontext — 编辑和图生图。

opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
tierselect质量 · 默认值:pro · 选项:pro、max
image_urlreq.text图像(URL)
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • tier: pro | max (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-kontext","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","tier":"pro","image_url":"https://example.com/input.jpg"}}'
flux-2

Flux 2 — 新一代 FLUX。支持文生图和图生图。

变体(addressAs)
参数类型描述
flux-2pixelPixel · 可用
flux-2:flowflowFlow · 不可用:flux-2:flow 路由仅支持文生图。请使用 flux-2 (Pixel) 进行图生图。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
tierselect版本 · 默认:pro · 选项:pro、flex
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
image_urltext图像(URL,用于图像到图像)
input_urlsarray图像(1–8 个 URL,用于图像到图像)
  • variants: flux-2, flux-2:flow
  • 层级:Pro · 1K、Pro · 2K、Flex · 1K、Flex · 2K
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • tier: pro | flex
  • mode: text-to-image | image-to-image (affects price)
  • resolution: 1K | 2K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-2","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","tier":"pro","mode":"text-to-image","resolution":"1K","input_urls":["https://example.com/input.jpg"]}}'
flux-3-image

FLUX 3 Image(Black Forest Labs)——文生图和参考图编辑(最多 10 张图)。分辨率为 768sq、1K、1.5K、2K 或 4K;每张图的价格取决于分辨率。

变体(addressAs)
参数类型描述
flux-3-image:whitewhiteWhite · Available
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: auto · options: auto, 21:9, 2:1, 16:9, 3:2, 7:5, 4:3, 5:4, 1:1, 4:5, 3:4, 5:7, 2:3, 9:16, 1:2, 9:21
resolutionselectResolution · default: 1K · options: 768sq, 1K, 1.5K, 2K, 4K
image_urltext参考图(URL,最多 10 张)
input_urlsarray参考图(URL,最多 10 张)
  • variants: flux-3-image:white
  • aspect_ratio: auto | 21:9 | 2:1 | 16:9 | 3:2 | 7:5 | 4:3 | 5:4 | 1:1 | 4:5 | 3:4 | 5:7 | 2:3 | 9:16 | 1:2 | 9:21
  • resolution: 768sq | 1K | 1.5K | 2K | 4K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-3-image","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"auto","resolution":"1K","input_urls":["https://example.com/input.jpg"]}}'
krea-2

Krea 2:Krea 的照片级真实感图像模型,支持风格控制。提供三个等级:Turbo(最快且最便宜)、Medium 和 Large(旗舰级,照片级真实感最强)。每张图像采用固定价格;宽高比不会改变价格。

变体(addressAs)
参数类型描述
krea-2flowFlow · 可用
opts 参数
参数类型描述
tierselect层级 · 默认值:turbo · 选项:turbo, medium, large
style_referenceselect风格参考 · 默认值:none · 可选值:none、image
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 4:3, 3:2, 16:9, 2.35:1, 4:5, 2:3, 9:16
creativityselect提示词解读自由度 · 默认值:medium · 选项:raw、low、medium、high
image_urltext风格参考图像(URL)
seednumberSeed(可选)
  • variants: krea-2
  • 层级:Turbo、Turbo · 参考图、Medium、Medium · 参考图、Large、Large · 参考图
  • tier: turbo | medium | large
  • style_reference: none | image
  • aspect_ratio: 1:1 | 4:3 | 3:2 | 16:9 | 2.35:1 | 4:5 | 2:3 | 9:16
  • creativity: raw | low | medium | high

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"krea-2","prompt":"A beautiful landscape in a digital illustration style.","opts":{"tier":"turbo","style_reference":"none","aspect_ratio":"1:1","creativity":"medium"}}'
flux-1.1-pro

FLUX 1.1 [pro] — Black Forest Labs 通过 Canvas 提供的旗舰模型。文生图:逼真度高,严格遵循提示词,比 FLUX.1 [pro] 快 6 倍。

变体(addressAs)
参数类型描述
flux-1.1-procanvasCanvas · 可用
flux-1.1-pro:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
seednumberSeed(可选)
  • variants: flux-1.1-pro, flux-1.1-pro:amber
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-1.1-pro","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1"}}'
flux-fill

FLUX.1 Fill — Black Forest Labs 的图像修复模型:重绘图像中由蒙版标记的区域(移除对象、替换背景、扩展场景)。需要源图像和黑白蒙版。

变体(addressAs)
参数类型描述
flux-fillcanvasCanvas · 可用
opts 参数
参数类型描述
tierselect版本 · 默认:dev · 选项:dev、pro
image_urlreq.text图像(URL)
mask_urlreq.text蒙版(URL,黑白图像,尺寸相同)
seednumberSeed(可选)
  • 变体:flux-fill
  • tier: dev | pro (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-fill","prompt":"A beautiful landscape in a digital illustration style.","opts":{"tier":"dev","image_url":"https://example.com/input.jpg","mask_url":"https://example.com/mask.png"}}'
seedream

ByteDance Seedream,图像生成和编辑。

opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
versionselect版本 · 默认:v4 · 选项:v3、v4
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
image_urltext图像(URL,用于 i2i,可选)
input_urlsarrayImages (1–10 URLs, for i2i, V4)
guidanceselectGuidance (V3) · default: 2.5 · options: 2.5, 1, 5, 7.5, 10
image_resolutionselectResolution (V4) · default: 1K · options: 1K, 2K, 4K
max_imagesselectMax images per set (V4) · default: 1 · options: 1, 2, 3, 4, 5, 6
seednumberSeed(可选)
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • version: v3 | v4 (affects price)
  • mode: text-to-image | image-to-image (affects price)
  • guidance: 2.5 | 1 | 5 | 7.5 | 10
  • image_resolution: 1K | 2K | 4K
  • max_images: 1 | 2 | 3 | 4 | 5 | 6

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedream","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","version":"v4","mode":"text-to-image","input_urls":["https://example.com/input.jpg"],"guidance":"2.5","image_resolution":"1K","max_images":"1"}}'
seedream-5

ByteDance Seedream 5,Lite 和 Pro,支持生成和编辑(seedream/5-*-text-to-image 和 seedream/5-*-image-to-image)。

opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
tierselect质量 · 默认值:lite · 选项:lite、pro
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
qualityselect输出质量 · 默认值:basic · 选项:basic、high、ultra
image_urltext图像(URL,用于 i2i)
input_urlsarray图像(1–14 个 URL,用于 i2i)
  • 层级:Lite、Pro · basic(1K)、Pro · high(2K)
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • tier: lite | pro
  • mode: text-to-image | image-to-image (affects price)
  • quality: basic | high | ultra

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedream-5","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","tier":"lite","mode":"text-to-image","quality":"basic","input_urls":["https://example.com/input.jpg"]}}'
seedream-layer

ByteDance Seedream 5.0 Pro,图层分解:将图像拆分为独立图层。

opts 参数
参数类型描述
image_urlreq.text图像(URL)
sizeselect分辨率 · 默认值:1.5K · 选项:auto、1K、1.5K、2K
output_formatselect格式 · 默认值:jpeg · 选项:jpeg、png
prompttext提示词(要突出显示的内容,使用 <bbox>)
  • tiers: auto, 1K, 1.5K, 2K
  • size: auto | 1K | 1.5K | 2K (affects price)
  • output_format: jpeg | png

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedream-layer","prompt":"A beautiful landscape in a digital illustration style.","opts":{"image_url":"https://example.com/input.jpg","size":"1.5K","output_format":"jpeg"}}'
seedream-4-5

ByteDance Seedream 4.5,高质量生成和编辑。

变体(addressAs)
参数类型描述
seedream-4-5pixelPixel · 可用
seedream-4-5:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
qualityselect输出质量 · 默认值:basic · 选项:basic、high
image_urltext图像(URL,用于 i2i)
input_urlsarray图像(1–14 个 URL,用于 i2i)
  • variants: seedream-4-5, seedream-4-5:amber
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • mode: text-to-image | image-to-image (affects price)
  • quality: basic | high

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedream-4-5","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","mode":"text-to-image","quality":"basic","input_urls":["https://example.com/input.jpg"]}}'
qwen-image

Alibaba Qwen Image — 图像生成和编辑(V1 和 V2)。

opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
versionselect版本 · 默认:v2 · 选项:v2、v1
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
image_urltext图像(URL,用于 i2i)
negative_prompttext负面提示词(V1)
strengthnumberStrength (0–1, for V1 i2i)
guidancetext引导强度(V1)
seednumberSeed(可选)
num_inference_stepsselectInference steps (V1) · default: 30 · options: 30, 20, 40, 50
accelerationselect加速(V1)· 默认值:none · 可选值:none、regular、high
  • tiers: V2 t2i, V2 i2i, V1 t2i (per megapixel), V1 i2i (per megapixel)
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • version: v2 | v1 (affects price)
  • mode: text-to-image | image-to-image (affects price)
  • num_inference_steps: 30 | 20 | 40 | 50
  • acceleration: none | regular | high

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen-image","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","version":"v2","mode":"text-to-image","num_inference_steps":"30","acceleration":"none"}}'
qwen3-image

Alibaba Qwen Image 3.0 — 图像生成和编辑。

变体(addressAs)
参数类型描述
qwen3-image:whitewhiteWhite · Available
qwen3-imagepixelPixel · 可用
qwen3-image:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
image_urltext图像(URL,用于 i2i)
input_urlsarray图像(1–3 个 URL,用于 i2i)
negative_prompttext负面提示词
seednumberSeed(可选)
  • variants: qwen3-image:white, qwen3-image, qwen3-image:amber
  • tiers: 1K, 2K
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • resolution: 1K | 2K
  • mode: text-to-image | image-to-image (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3-image","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1K","mode":"text-to-image","input_urls":["https://example.com/input.jpg"]}}'
qwen3-pro

Alibaba Qwen Image 3.0 Pro — 图像生成和编辑。

变体(addressAs)
参数类型描述
qwen3-propixelPixel · 可用
qwen3-pro:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
image_urltext图像(URL,用于 i2i)
input_urlsarray图像(1–3 个 URL,用于 i2i)
negative_prompttext负面提示词
seednumberSeed(可选)
  • variants: qwen3-pro, qwen3-pro:amber
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • resolution: 1K | 2K (affects price)
  • mode: text-to-image | image-to-image (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3-pro","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1K","mode":"text-to-image","input_urls":["https://example.com/input.jpg"]}}'
qwen-image-2-1

Alibaba Qwen Image 2.1 — 在一个模型中实现图像生成和编辑(docs.Pixel/market/qwen2-1/*)。最多支持 10 张参考图像、基于蒙版的局部编辑,以及原生透明(alpha)输出。

变体(addressAs)
参数类型描述
qwen-image-2-1pixelPixel · 可用
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 4:3, 3:4, 3:2, 2:3, 16:9, 9:16, auto
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
image_urltext图像(URL,用于 i2i)
input_urlsarray图像(1–10 个 URL,用于 i2i)
mask_urltext蒙版(URL;白色 = 更改,黑色 = 保留)
backgroundselect背景 · 默认值:opaque · 选项:opaque、transparent
output_formatselect输出格式 · 默认值:png · 选项:png、webp、jpeg
enhance_promptbool增强提示词 · 默认值:false
seednumberSeed(可选)
  • variants: qwen-image-2-1
  • tiers: 1K, 2K
  • aspect_ratio: 1:1 | 4:3 | 3:4 | 3:2 | 2:3 | 16:9 | 9:16 | auto
  • resolution: 1K | 2K
  • mode: text-to-image | image-to-image (affects price)
  • background: opaque | transparent
  • output_format: png | webp | jpeg

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen-image-2-1","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1K","mode":"text-to-image","input_urls":["https://example.com/input.jpg"],"background":"opaque","output_format":"png"}}'
recraft

Recraft — 移除背景并提升图片分辨率。

opts 参数
参数类型描述
modeselect模式 · 默认值:remove-bg · 选项:remove-bg、upscale
image_urlreq.text图像(URL)
  • mode: remove-bg | upscale (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"recraft","prompt":"A beautiful landscape in a digital illustration style.","opts":{"mode":"remove-bg","image_url":"https://example.com/input.jpg"}}'
image-upscale

图像放大 — 将图像放大 ×1、×2、×4 或 ×8(默认为 ×2)。发送一个可公开访问的 image_url(JPG、PNG、WEBP、HEIC、BMP 或 GIF,最大 20 MB);输出尺寸严格为原图尺寸 × 放大倍数,绝不裁剪。×1 保持尺寸不变,仅提升画质。按每张图像的输出总像素数计费:不超过 4.19 MP — 1.5 积分,不超过 16.78 MP — 4 积分,不超过 67.11 MP — 12 积分;使用 ×1 时,计费档位取决于原图尺寸。输出超过 67.11 MP 或任一边超过 16,384 px 的请求将在收费前被拒绝。服务器会在任务开始前测量原图尺寸。

opts 参数
参数类型描述
upscale_factorselectScale · default: 2 · options: 1, 2, 4, 8
image_urlreq.text图像(URL)
  • 档位:2K · ≤4.19 MP,4K · ≤16.78 MP,8K · ≤67.11 MP
  • upscale_factor: 1 | 2 | 4 | 8

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"image-upscale","prompt":"A beautiful landscape in a digital illustration style.","opts":{"upscale_factor":"2","image_url":"https://example.com/input.jpg"}}'
z-image

Z-Image — 超低成本图像生成。

opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"z-image","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1"}}'
z-image-turbo

Z-Image Turbo——快速、低成本的 1K 或 2K 文生图。智能提示词改写会使价格翻倍。

变体(addressAs)
参数类型描述
z-image-turbo:whitewhiteWhite · Available
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K
prompt_extendbool智能提示词改写(费用更高) · 默认值:false
  • variants: z-image-turbo:white
  • aspect_ratio: 1:1 | 4:3 | 3:4 | 16:9 | 9:16 | 3:2 | 2:3
  • resolution: 1K | 2K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"z-image-turbo","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1K"}}'
hunyuan-image-3-5

Hunyuan Image 3.5(Tencent)— 文生图和基于参考图像的生成(最多 5 张图像),支持 1K、2K 或 4K。按图像计费;4K 费用更高。

变体(addressAs)
参数类型描述
hunyuan-image-3-5:whitewhiteWhite · Available
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: auto, 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9
resolutionselect分辨率 · 默认值:1K · 选项:1K、2K、4K
image_urltext参考图像(URL,最多 5 张)
input_urlsarray参考图像(URL,最多 5 张)
  • variants: hunyuan-image-3-5:white
  • aspect_ratio: auto | 1:1 | 2:3 | 3:2 | 3:4 | 4:3 | 4:5 | 5:4 | 9:16 | 16:9 | 21:9
  • resolution: 1K | 2K | 4K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"hunyuan-image-3-5","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","resolution":"1K","input_urls":["https://example.com/input.jpg"]}}'
wan-2-7-image

Wan 2.7 Image:图像生成和编辑(Alibaba Wan)。不带 input_urls 时使用 T2I;通过 input_urls 使用 I2I(最多 9 个)。

变体(addressAs)
参数类型描述
wan-2-7-imagepixelPixel · 可用
wan-2-7-image:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 16:9, 9:16, 4:3, 3:4
tierselect质量 · 默认值:standard · 选项:standard、pro
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
enable_sequentialbool顺序生成(最多 12 张图片)· 默认值:false
resolutionselect分辨率 · 默认值:2K · 选项:1K、2K、4K
nselectNumber of images · default: 1 · options: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
seednumberSeed(0–2147483647,可选)
watermarkbool水印 · 默认值:false
thinking_modeselect思考模式 · 默认值:false · 选项:false, true
color_palettearray颜色调色板(3–10 种颜色,仅当 enable_sequential=false 时)
bbox_listarray边界框区域(每个 input_urls 图像对应一个列表,最多 2 个框)
image_urltext图像(URL,用于 i2i)
input_urlsarray图像(1–9 个 URL,用于 i2i)
  • variants: wan-2-7-image, wan-2-7-image:white
  • aspect_ratio: 1:1 | 16:9 | 9:16 | 4:3 | 3:4
  • tier: standard | pro (affects price)
  • mode: text-to-image | image-to-image (affects price)
  • resolution: 1K | 2K | 4K
  • n: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12
  • thinking_mode: false | true

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan-2-7-image","prompt":"A beautiful landscape in a digital illustration style.","opts":{"aspect_ratio":"1:1","tier":"standard","mode":"text-to-image","resolution":"2K","n":"1","thinking_mode":"false","input_urls":["https://example.com/input.jpg"]}}'
grok-imagine-image-2-0

Grok Imagine Image 2.0 (xAI) — 图像生成和基于参考图像的编辑。两种模式下每张图像均统一收取 4 Pixel credits;编辑模式接受 1–5 个参考图像 URL。

opts 参数
参数类型描述
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-edit
aspect_ratioselectAspect ratio · default: 1:1 · options: 1:1, 2:3, 3:2, 16:9, 9:16
image_urlsarray参考图(1–5 个 URL,用于编辑)
image_urltext图像(URL,用于编辑)
  • 层级:生成、编辑
  • mode: text-to-image | image-edit
  • aspect_ratio: 1:1 | 2:3 | 3:2 | 16:9 | 9:16

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-imagine-image-2-0","prompt":"A beautiful landscape in a digital illustration style.","opts":{"mode":"text-to-image","aspect_ratio":"1:1","image_urls":["https://example.com/input.jpg"]}}'
minimax-h3-image

MiniMax H3 Image — 图像生成与编辑。支持 1K/2K/4K(最高约 16 MP)或自定义宽度×高度(0.66–20 MP,每边最高 8192 px),支持 10 种宽高比及 auto。编辑时可接受 1–16 张参考图像;每张参考图像会使价格增加 0.5 积分。默认开启提示词扩展;输出格式为 PNG 或 JPEG;可选 seed。

opts 参数
参数类型描述
modeselect模式 · 默认值:text-to-image · 选项:text-to-image、image-to-image
aspect_ratioselectAspect ratio · default: auto · options: auto, 1:1, 5:4, 4:3, 3:2, 16:9, 21:9, 4:5, 3:4, 2:3, 9:16
resolutionselect分辨率 · 默认值:2K · 选项:1K、2K、4K
widthnumber自定义宽度,单位为 px(与高度一起设置)
heightnumber自定义高度,单位为 px(与宽度一起设置)
expand_promptbool增强提示词(默认开启)· 默认值:true
output_formatselect格式 · 默认值:png · 选项:png、jpeg
seednumberSeed(可选)
image_urltext图像(URL,参考图)
image_urlsarray图像(1–16 个 URL,作为参考)
  • tiers: 1K, 2K, 4K
  • mode: text-to-image | image-to-image
  • aspect_ratio: auto | 1:1 | 5:4 | 4:3 | 3:2 | 16:9 | 21:9 | 4:5 | 3:4 | 2:3 | 9:16
  • resolution: 1K | 2K | 4K
  • output_format: png | jpeg

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-image","prompt":"A beautiful landscape in a digital illustration style.","opts":{"mode":"text-to-image","aspect_ratio":"auto","resolution":"2K","output_format":"png"}}'
topaz-upscale-generative

通过 Flow 使用 Topaz Generative Upscale:按已开始处理的百万像素计费,具体取决于所选 Topaz 模型和输出尺寸;必须提供源图像和输出图像的百万像素数。

变体(addressAs)
参数类型描述
topaz-upscale-generativeflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
modelselectTopaz 模型 · 默认值:Wonder 3 · 选项:Wonder, Wonder 2, Wonder 3, Wonder 3.5, Recover 3, Standard MAX, Redefine, Recovery, Recovery V2
source_megapixelsreq.number源尺寸,MP
output_megapixelsreq.number输出大小,MP
upscale_factorselect倍数 · 默认值:2 · 选项:2、4
  • 变体:topaz-upscale-generative
  • model: Wonder | Wonder 2 | Wonder 3 | Wonder 3.5 | Recover 3 | Standard MAX | Redefine | Recovery | Recovery V2
  • upscale_factor: 2 | 4

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"topaz-upscale-generative","prompt":"A beautiful landscape in a digital illustration style.","opts":{"image_url":"https://example.com/input.jpg","model":"Wonder 3","source_megapixels":1,"output_megapixels":4,"upscale_factor":"2"}}'
topaz-sharpen

通过 Flow 使用 Topaz Sharpen:按已开始处理的百万像素计费,经典模型按 24MP 桶计费,Super Focus 模型按 20MP 桶计费。

变体(addressAs)
参数类型描述
topaz-sharpenflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
modelselectTopaz 模型 · 默认值:Standard · 选项:Standard, Strong, Lens Blur V2, Motion Blur, Natural, Refocus, Wildlife, Portrait, Auto Sharpen, Super Focus V2, Super Focus V3
source_megapixelsreq.numberSize, MP
  • 变体:topaz-sharpen
  • model: Standard | Strong | Lens Blur V2 | Motion Blur | Natural | Refocus | Wildlife | Portrait | Auto Sharpen | Super Focus V2 | Super Focus V3

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"topaz-sharpen","prompt":"A beautiful landscape in a digital illustration style.","opts":{"image_url":"https://example.com/input.jpg","model":"Standard","source_megapixels":1}}'
topaz-denoise

通过 Flow 使用 Topaz Denoise:按已开始处理的百万像素计费,Normal/Strong/Extreme 按 24MP 桶计费,Denoise Max 按 20MP 桶计费。

变体(addressAs)
参数类型描述
topaz-denoiseflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
modelselectTopaz 模型 · 默认值:Normal · 选项:Normal, Strong, Extreme, Denoise Max
source_megapixelsreq.numberSize, MP
  • 变体:topaz-denoise
  • model: Normal | Strong | Extreme | Denoise Max

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"topaz-denoise","prompt":"A beautiful landscape in a digital illustration style.","opts":{"image_url":"https://example.com/input.jpg","model":"Normal","source_megapixels":1}}'
topaz-upscale-precision

通过 Flow 使用 Topaz Precision Upscale(Standard V2、High Fidelity V2/V3、Low Resolution V2、CGI、Text Refine、Faces)——忠实放大,不生成额外细节,输出最高可达 512 MP,适用于打印。按输出每 24 MP 计费,不足 24 MP 按 24 MP 计;source_megapixels 和 output_megapixels 为必填项,upscale_factor 为 1–4 的放大倍数,可为小数。

变体(addressAs)
参数类型描述
topaz-upscale-precisionflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
modelselectTopaz 模型 · 默认值:Standard V2 · 选项:Standard V2、High Fidelity V2、High Fidelity V3、Low Resolution V2、CGI、Text Refine、Faces
source_megapixelsreq.number源尺寸,MP
output_megapixelsreq.number输出大小,MP
upscale_factornumber放大倍数(1–4,可为小数)· 默认值:2
output_formatselect格式 · 默认值:jpeg · 选项:jpeg、png
face_enhancementbool面部增强 · 默认值:true
  • 变体:topaz-upscale-precision
  • model: Standard V2 | High Fidelity V2 | High Fidelity V3 | Low Resolution V2 | CGI | Text Refine | Faces
  • output_format: jpeg | png

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"topaz-upscale-precision","prompt":"A beautiful landscape in a digital illustration style.","opts":{"image_url":"https://example.com/input.jpg","model":"Standard V2","source_megapixels":1,"output_megapixels":4,"upscale_factor":"2","output_format":"jpeg"}}'

Video (86)

model输入单位变体opts 参数
seedance-2.0-facetext, image, video每秒seedance-2.0-face:white (unavailable)duration, resolution, image_url, video_url
seedance-2.0-fast-facetext每秒seedance-2.0-fast-face:whiteduration, resolution
happyhorse-1.1text, image, video每秒happyhorse-1.1:white (unavailable)input, duration, resolution, image_url, video_url
kling-video-o1text, image每秒kling-video-o1:whiteduration, resolution, aspect_ratio, image_url
MiniMax-H3-Context-IRtext, image, video, audio每个片段MiniMax-H3-Context-IR:white (unavailable)input
MiniMax-H3-Regenerationtext, video每秒MiniMax-H3-Regeneration:white (unavailable)duration
MiniMax-Hailuo-02text, image每秒MiniMax-Hailuo-02:whiteduration, resolution, image_url
skyreels-v4-fasttext, image, audio, video每秒skyreels-v4-fast:whiteduration, resolution, aspect_ratio, image_url, reference_image_urls, audio_url, video_url, video_mode
skyreels-v4-stdtext, image, audio, video每秒skyreels-v4-std:whiteduration, resolution, aspect_ratio, image_url, reference_image_urls, audio_url, video_url, video_mode
wan2.7-r2vtext, image, video每秒wan2.7-r2v:white (unavailable)duration, resolution, image_url, video_url
wan2.7-videoedittext, video每秒wan2.7-videoedit:white (unavailable)duration, resolution, video_url
wan3.0-videotext, image, video每秒wan3.0-video:whiteduration, resolution, aspect_ratio, image_url
minimax-h3-max-stylestext, image每秒minimax-h3-max-stylesstyle, duration, aspect_ratio, damage_level, image_url, seed
minimax-h3-max-relightvideo, image每秒minimax-h3-max-relightvideo_url, reference_image_url, resolution, aspect_ratio, seed
minimax-h3-max-extendtext, video每秒minimax-h3-max-extendvideo_url, engine, duration, resolution, aspect_ratio, output, enable_prompt_expansion, seed
minimax-h3-max-inserttext, video每秒minimax-h3-max-insertvideo_url, start_time, resume_time, duration, resolution, color_match, enable_prompt_expansion, seed
minimax-h3-max-cameraimage, text每秒minimax-h3-max-cameraimage_url, camera_move, duration, resolution, prompt_expansion_mode, seed
runway-gen-4-5text, image每个片段runway-gen-4-5duration, aspect_ratio, image_url, seed
p-video-2text, image, audio每个片段p-video-2duration, resolution, draft, aspect_ratio, image_url, last_frame_url, audio_url, save_audio, seed
p-video-2-protext, image每个片段p-video-2-promode, duration, resolution, aspect_ratio, image_url, last_frame_url, seed
p-video-2-pro-referencetext, image, video, audio每个片段p-video-2-pro-referencemode, duration, resolution, aspect_ratio, reference_image_urls, reference_video_urls, reference_audio_urls, keep_audio, first_frame_url, last_frame_url, seed
p-video-editvideo, image, text每秒p-video-editvideo_url, input_urls, draft, seed
luma-ray-3-2text, image每个片段luma-ray-3-2resolution, duration, aspect_ratio, image_url, end_image_url, loop
veo-3.1text, image每个片段veo-3.1:white (unavailable), veo-3.1, veo-3.1:flowaspect_ratio, tier, generationType, resolution, duration, imageUrls
kling-v2.6text, image每个片段kling-v2.6, kling-v2.6:flow, kling-v2.6:prime, kling-v2.6:whiteaspect_ratio, duration, input, resolution, image_url, audio
kling-v3.0text, image每秒kling-v3.0, kling-v3.0:primeaspect_ratio, duration, mode, multi_shots, multi_prompt, audio, image_url, last_frame_url, kling_elements
kling-v3-omnitext, image每秒kling-v3-omni, kling-v3-omni:primeinput, aspect_ratio, resolution, audio, duration, image_url, last_frame_url
kling-v3-omni-referencetext, image, video每秒kling-v3-omni-referencevideo_input, aspect_ratio, resolution, audio, duration, image_url, input_urls, video_url
kling-v3-omni-transformationvideo, image每秒kling-v3-omni-transformationresolution, aspect_ratio, duration, video_url, image_url, input_urls
kling-v4text, image, video, audio每秒kling-v4input, aspect_ratio, resolution, duration, image_url, keyframe_urls, last_frame_url, reference_image_urls, reference_video_urls, keep_original_sound, elements, voices
kling-v4-flashtext, image, video, audio每秒kling-v4-flashinput, aspect_ratio, resolution, duration, image_url, keyframe_urls, last_frame_url, reference_image_urls, reference_video_urls, keep_original_sound, elements, voices
kling-v3-turbotext, image每秒—aspect_ratio, duration, input, resolution, image_url
kling-v2-5-turbotext, image每个片段—aspect_ratio, duration, input, image_url
kling-ai-avatarimage, audio每秒—tier, image_url, audio_url
kling-motion-controlimage, video每秒—version, resolution, duration, image_url, video_url
seedance-2.0text, image每秒seedance-2.0:white (unavailable), seedance-2.0:pixel, seedance-2.0:prime (unavailable)tier, input, video_input, resolution, duration, image_url, last_frame_url, input_urls, video_url, reference_audio_urls, web_search, audio
seedance-2.5text, image每秒seedance-2.5:white (unavailable), seedance-2.5:rep (unavailable), seedance-2.5:pixel, seedance-2.5:prime (unavailable)input, video_input, aspect_ratio, resolution, duration, image_url, last_frame_url, input_urls, video_url, reference_audio_urls, audio
seedance-2-0-minitext, image每秒seedance-2-0-mini:pixel, seedance-2-0-mini:primeresolution, input, video_input, duration, image_url, last_frame_url, audio
pixverse-v6text, image每秒pixverse-v6:pixel, pixverse-v6:whitemode, aspect_ratio, quality, duration, audio, multi_clip, seed, image_url, input_urls, last_frame_url, image_references, task_id, video_url
pixverse-v6-maxtext, image每秒pixverse-v6-max:pixel, pixverse-v6-max:flowmode, aspect_ratio, quality, duration, audio, seed, image_url, input_urls
wan-3-0text, image每秒wan-3-0, wan-3-0:amber (unavailable)mode, duration, resolution, aspect_ratio, image_url, negative_prompt, seed, prompt_extend
wan-3-0-primetext, image, video, audio每秒wan-3-0-prime:white (unavailable), wan-3-0-prime:pixel, wan-3-0-prime:flow (unavailable)mode, input, duration, resolution, aspect_ratio, audio, first_frame_url, last_frame_url, reference_image_urls, reference_video_urls, reference_audio_urls, reference_file_urls, reference_link_urls, seed
flux-3text, image, video每秒flux-3mode, duration, quality, resolution, aspect_ratio, image_url, video_url, generate_audio
wan-2-7text, image, video每秒wan-2-7:white, wan-2-7:pixelaspect_ratio, duration, mode, resolution, image_url, last_frame_url, first_clip_url, video_url, audio_url, first_frame, reference_voice, negative_prompt, seed, prompt_extend, watermark, audio_setting
wan-2-6text, image每个片段wan-2-6, wan-2-6:amber (unavailable)input, duration, resolution, image_url
wan-2-5text, image每个片段—input, duration, aspect_ratio, resolution, image_url, negative_prompt, enable_prompt_expansion, seed, nsfw_checker
wan-2-2text, image每个片段—aspect_ratio, input, resolution, image_url
wan-speech-to-videotext, image, audio每秒wan-speech-to-video:flowresolution, image_url, audio_url, num_frames, frames_per_second, negative_prompt, num_inference_steps, guidance_scale, shift
hailuo-2-3image每个片段hailuo-2-3, hailuo-2-3:whitetier, duration, resolution, image_url
vidu-q3text, image每秒vidu-q3:white, vidu-q3, vidu-q3:amber (unavailable)input, duration, resolution, image_url
vidu-q2text, image每秒vidu-q2tier, input, duration, resolution, image_url
vidu-q4text, image, audio每秒vidu-q4:pixel, vidu-q4:flowinput, resolution, duration, audio, aspect_ratio, image_url, reference_image_urls, reference_audio_urls
happyhorsetext, image, video每秒happyhorse:pixel, happyhorse:prime (unavailable)aspect_ratio, duration, version, input, resolution, image_url, reference_image, video_url, audio_setting, seed
grok-imagine-videotext, image每秒grok-imagine-video, grok-imagine-video:white (unavailable)version, input, mode, aspect_ratio, resolution, duration, image_url, first_frame_url, last_frame_url, loop_url, keyframes
grok-imagine-upscalevideo每次放大—task_id, resolution
grok-imagine-extendtext, video每个片段—task_id, extend_times, source_resolution, extend_at
minimax-h3-maxtext, image每秒minimax-h3-maxinput, resolution, duration, aspect_ratio, image_url, last_frame_url, target_audio_url, prompt_expansion_mode, seed, enable_safety_checker
minimax-h3-max-turbotext, image每秒minimax-h3-max-turboinput, resolution, duration, aspect_ratio, image_url, last_frame_url, target_audio_url, prompt_expansion_mode, seed, enable_safety_checker
gemini-omni-flash-1-1-flowtext, image, video每秒gemini-omni-flash-1-1-flowinput, resolution, aspect_ratio, image_url, video_url, end_image_url, image_urls, reference_video_urls, source_duration_seconds, duration
minimax-h3text, image, video每秒minimax-h3, minimax-h3:white, minimax-h3:amber (unavailable)input, aspect_ratio, resolution, duration, first_frame_url, last_frame_url, reference_image_urls, reference_video_urls, reference_audio_urls, input_video_duration
minimax-h3-litetext, image, video, audio每秒minimax-h3-lite, minimax-h3-lite:amber (unavailable)input, aspect_ratio, resolution, duration, first_frame_url, last_frame_url, reference_image_urls, reference_video_urls, reference_audio_urls
minimax-h3-highspeedtext, image, video, audio每秒minimax-h3-highspeedinput, aspect_ratio, resolution, duration, first_frame_url, last_frame_url, reference_image_urls, reference_video_urls, reference_audio_urls
gemini-omni-videotext, image每个片段—input, aspect_ratio, duration, resolution, image_url, audio_ids, video_list, seed
gemini-omni-flash-1-1text, image, video, audio每个片段—input, aspect_ratio, duration, resolution, image_url, audio_ids, video_list, seed
gemini-omni-flash-1-2text, image每秒gemini-omni-flash-1-2, gemini-omni-flash-1-2:whiteinput, aspect_ratio, duration, resolution, image_url, last_frame_url, image_urls, character_ids
gemini-omni-flash-1-2-extendtext, image, video每秒—source_task_id, duration, resolution, image_urls
runway-gen4text, image每个片段—aspect_ratio, duration, quality, image_url
omnihuman-1-5image, audio, text每秒—image_url, audio_url, output_resolution, pe_fast_mode, seed, mask_url
omnihuman-2-0image, audio, text每秒—image_url, audio_url, enable_watermark
volcengine-lipsyncvideo, audio每秒—mode, video_url, audio_url, separate_vocal, open_scenedet, align_audio
topaz-video-upscalevideo每秒topaz-video-upscalevideo_url, upscale_factor, model, target_fps
video-upscalevideo每秒—video_url, resolution
omnihuman-v2image, audio, text每秒omnihuman-v2:flowimage_url, audio_url
omnihuman-v1.5image, audio, text每秒omnihuman-v1.5:flowimage_url, audio_url, mask_url, resolution, turbo_mode
lipsync-2-provideo, audio每秒lipsync-2-pro:flowvideo_url, audio_url, sync_mode
sync-3video, audio每秒sync-3:flowvideo_url, audio_url, sync_mode
react-1video, audio每秒react-1:flowvideo_url, audio_url, emotion, model_mode, temperature, sync_mode
minimax-h3-max-lipsyncimage, audio每秒minimax-h3-max-lipsync:flowimage_url, audio_url, resolution, enable_transcription, seed
sync-3-avatarimage, audio每秒sync-3-avatar:flowimage_url, audio_url
heygen-avatar-4image, audio每秒heygen-avatar-4:flowimage_url, audio_url, resolution, aspect_ratio, talking_style
heygen-avatar-vtext, audio每秒heygen-avatar-v:flowavatar, audio_url, resolution, aspect_ratio
veed-fabric-1.0image, audio每秒veed-fabric-1.0:flowimage_url, audio_url, resolution
ltx-2.5-audio-to-videoimage, audio, text每秒ltx-2.5-audio-to-video:flowimage_url, audio_url, aspect_ratio, guidance_scale
pikaimage, text每个片段pikaimage_url, resolution, seed, negative_prompt, duration
p-video-avatartext, image, audio每秒p-video-avatarimage_url, audio_url, voice_script, resolution
suno-music-videoaudio每个片段suno-music-video, suno-music-video:whitesource_task_id, source_track, source_audio_id, author, domain_name
seedance-2.0-face

Seedance 2.0 Face — 暂时不可用。

变体(addressAs)
参数类型描述
seedance-2.0-face:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
durationselect时长,秒 · 默认值:5 · 选项:5
resolutionselect分辨率 · 默认值:480P · 选项:480P、720P、1080P
image_urltext图像(URL)
video_urltext参考视频(URL)
  • variants: seedance-2.0-face:white
  • duration: 5
  • resolution: 480P | 720P | 1080P

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedance-2.0-face","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"480P"}}'
seedance-2.0-fast-face

Seedance 2.0 Fast Face(ByteDance)— 支持 480P 或 720P 的文生视频,无音频。按秒计费。

变体(addressAs)
参数类型描述
seedance-2.0-fast-face:whitewhiteWhite · Available
opts 参数
参数类型描述
durationselect时长,秒 · 默认值:5 · 选项:5
resolutionselect分辨率 · 默认值:480P · 选项:480P、720P
  • variants: seedance-2.0-fast-face:white
  • duration: 5
  • resolution: 480P | 720P

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedance-2.0-fast-face","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"480P"}}'
happyhorse-1.1

HappyHorse 1.1 — 暂时不可用。

变体(addressAs)
参数类型描述
happyhorse-1.1:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video
durationselect时长,秒 · 默认值:5 · 选项:5、10
resolutionselect分辨率 · 默认值:720P · 选项:720P、1080P
image_urltext图像(URL)
video_urltext参考视频(URL)
  • variants: happyhorse-1.1:white
  • input: text-to-video | image-to-video | reference-to-video
  • duration: 5 | 10
  • resolution: 720P | 1080P

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"happyhorse-1.1","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","duration":"5","resolution":"720P"}}'
kling-video-o1

Kling Video O1:文生视频或首帧视频。支持 720P(Standard)或 1080P(Pro),时长为 5 或 10 秒。按分辨率按秒计费。

变体(addressAs)
参数类型描述
kling-video-o1:whitewhiteWhite · Available
opts 参数
参数类型描述
durationselect时长,秒 · 默认值:5 · 选项:5、10
resolutionselect分辨率 · 默认值:720P · 选项:720P、1080P
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
image_urltext图像(URL,用于 I2V)
  • variants: kling-video-o1:white
  • duration: 5 | 10
  • resolution: 720P | 1080P
  • aspect_ratio: 16:9 | 9:16 | 1:1

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-video-o1","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"720P","aspect_ratio":"16:9"}}'
MiniMax-H3-Context-IR

MiniMax H3 Context IR:暂时不可用。

变体(addressAs)
参数类型描述
MiniMax-H3-Context-IR:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
inputselect输入 · 默认值:context · 选项:context
  • variants: MiniMax-H3-Context-IR:white
  • input:上下文

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"MiniMax-H3-Context-IR","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"context"}}'
MiniMax-H3-Regeneration

MiniMax H3 Regeneration:暂时不可用。

变体(addressAs)
参数类型描述
MiniMax-H3-Regeneration:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
durationnumber时长,秒 · 默认值:5
  • variants: MiniMax-H3-Regeneration:white

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"MiniMax-H3-Regeneration","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5"}}'
MiniMax-Hailuo-02

MiniMax Hailuo 02:文本转视频或首帧视频。512P(仅首帧)、768P 或 1080P(仅 5 秒)。按分辨率以秒数计费。

变体(addressAs)
参数类型描述
MiniMax-Hailuo-02:whitewhiteWhite · Available
opts 参数
参数类型描述
durationselect时长,秒 · 默认值:5 · 选项:5、10
resolutionselect分辨率 · 默认值:768P · 选项:512P、768P、1080P
image_urltext图像(URL,用于 I2V)
  • variants: MiniMax-Hailuo-02:white
  • duration: 5 | 10
  • resolution: 512P | 768P | 1080P

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"MiniMax-Hailuo-02","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"768P"}}'
skyreels-v4-fast

SkyReels V4 Fast — 支持文生视频、首帧视频或会说话的头像:角色照片加上语音片段,可生成带身体动作的唇形同步语音。视频参考:重复片段中的动作或延长片段。支持 480P / 720P / 1080P,3–15 秒。按分辨率逐秒计费;视频参考有单独的费率,在动作参考模式下,输出时长跟随片段长度(最长 10 s),并按 10 s 计费。

变体(addressAs)
参数类型描述
skyreels-v4-fast:whitewhiteWhite · Available
opts 参数
参数类型描述
durationselectDuration, sec · default: 5 · options: 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
resolutionselect分辨率 · 默认值:480P · 选项:480P、720P、1080P
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 4:3, 1:1, 9:16, 3:4
image_urltext图像(URL,用于 I2V)
reference_image_urlsarray角色照片(1–5 个 URL,用于头像)
audio_urltext头像语音(音频 URL,最长 15 s)
video_urltext参考视频(URL,MP4/MOV,最长 15 s)
video_modeselect参考视频模式 · 默认值:reference · 选项:reference、extend
  • variants: skyreels-v4-fast:white
  • duration: 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15
  • resolution: 480P | 720P | 1080P
  • aspect_ratio: 16:9 | 4:3 | 1:1 | 9:16 | 3:4
  • video_mode: reference | extend

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"skyreels-v4-fast","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"480P","aspect_ratio":"16:9","reference_image_urls":["https://example.com/input.jpg"],"video_mode":"reference"}}'
skyreels-v4-std

SkyReels V4 Standard — 更高质量的 SkyReels V4 版本:支持文生视频、首帧视频或会说话的头像(角色照片加上语音片段)。视频参考:重复片段中的动作或延长片段。支持 480P / 720P / 1080P,3–15 秒。按分辨率逐秒计费;视频参考有单独的费率,在动作参考模式下,输出时长跟随片段长度(最长 10 s),并按 10 s 计费。

变体(addressAs)
参数类型描述
skyreels-v4-std:whitewhiteWhite · Available
opts 参数
参数类型描述
durationselectDuration, sec · default: 5 · options: 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
resolutionselect分辨率 · 默认值:480P · 选项:480P、720P、1080P
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 4:3, 1:1, 9:16, 3:4
image_urltext图像(URL,用于 I2V)
reference_image_urlsarray角色照片(1–5 个 URL,用于头像)
audio_urltext头像语音(音频 URL,最长 15 s)
video_urltext参考视频(URL,MP4/MOV,最长 15 s)
video_modeselect参考视频模式 · 默认值:reference · 选项:reference、extend
  • variants: skyreels-v4-std:white
  • duration: 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15
  • resolution: 480P | 720P | 1080P
  • aspect_ratio: 16:9 | 4:3 | 1:1 | 9:16 | 3:4
  • video_mode: reference | extend

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"skyreels-v4-std","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"480P","aspect_ratio":"16:9","reference_image_urls":["https://example.com/input.jpg"],"video_mode":"reference"}}'
wan2.7-r2v

Wan 2.7 Reference-to-Video(Alibaba):根据参考图像和/或视频生成视频(使用视频时最长 10 秒)。根据分辨率按秒计费。

变体(addressAs)
参数类型描述
wan2.7-r2v:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
durationnumber时长,秒 · 默认值:5
resolutionselect分辨率 · 默认值:720P · 选项:720P、1080P
image_urltext参考图片(URL)
video_urltext参考视频(URL)
  • variants: wan2.7-r2v:white
  • resolution: 720P | 1080P

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan2.7-r2v","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"720P"}}'
wan2.7-videoedit

Wan 2.7 Video Edit:暂时不可用。

变体(addressAs)
参数类型描述
wan2.7-videoedit:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
durationnumber时长,秒 · 默认值:5
resolutionselect分辨率 · 默认值:720P · 选项:720P、1080P
video_urlreq.text视频(URL)
  • variants: wan2.7-videoedit:white
  • resolution: 720P | 1080P

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan2.7-videoedit","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"720P","video_url":"https://example.com/input.mp4"}}'
wan3.0-video

Wan 3.0 Video(Alibaba):以文本生成视频,或生成以首帧开始的视频,支持 480P / 720P / 1080P。根据分辨率按秒计费。

变体(addressAs)
参数类型描述
wan3.0-video:whitewhiteWhite · Available
opts 参数
参数类型描述
durationnumber时长,秒 · 默认值:5
resolutionselect分辨率 · 默认值:480P · 选项:480P、720P、1080P
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 4:3, 1:1, 3:4, 9:16
image_urltext图像(URL,用于 I2V)
  • variants: wan3.0-video:white
  • resolution: 480P | 720P | 1080P
  • aspect_ratio: 16:9 | 4:3 | 1:1 | 3:4 | 9:16

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan3.0-video","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"480P","aspect_ratio":"16:9"}}'
minimax-h3-max-styles

MiniMax H3 Max Styles — 使用现成风格生成带声音的视频:VHS 磁带、70 年代卡通、低多边形、手绘动画或 16 位像素艺术。支持从文本或首帧(image_url)生成,768p,5–15 秒。按秒计费。

变体(addressAs)
参数类型描述
minimax-h3-max-stylesflowFlow · 可用
opts 参数
参数类型描述
styleselect风格 · 默认值:vhs · 选项:vhs、retro-toon-70s、low-poly、hand-drawn、16bit-pixel
durationselectDuration, sec · default: 5 · options: 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 21:9, 4:3, 1:1, 3:4, 9:16
damage_levelselect磁带磨损(仅限 VHS)· 默认值:medium · 选项:light、medium、heavy
image_urltext第一帧(URL,可选)
seednumberSeed(可选)
  • variants: minimax-h3-max-styles
  • style: vhs | retro-toon-70s | low-poly | hand-drawn | 16bit-pixel
  • duration: 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15
  • aspect_ratio: 16:9 | 21:9 | 4:3 | 1:1 | 3:4 | 9:16
  • damage_level: light | medium | heavy

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-max-styles","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"style":"vhs","duration":"5","aspect_ratio":"16:9","damage_level":"medium"}}'
minimax-h3-max-relight

MiniMax H3 Max Relight — 根据光照参考为现有视频(最长 15 秒)重新布光:光照参考是一张按照所需方式为球体打光的图像。场景和动作保持不变,仅改变光线。480p / 768p / 1080p / 2K。按源视频的秒数计费。

变体(addressAs)
参数类型描述
minimax-h3-max-relightflowFlow · 可用
opts 参数
参数类型描述
video_urlreq.text视频(URL)
reference_image_urlreq.text光照参考(球体渲染图的 URL)
resolutionselectResolution · default: 768p · options: 480p, 768p, 1080p, 2k
aspect_ratioselect宽高比 · 默认值:adaptive · 选项:adaptive、21:9、16:9、4:3、1:1、3:4、9:16
seednumberSeed(可选)
  • variants: minimax-h3-max-relight
  • resolution: 480p | 768p | 1080p | 2k
  • aspect_ratio: adaptive | 21:9 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-max-relight","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_url":"https://example.com/input.mp4","reference_image_url":"https://example.com/input.jpg","resolution":"768p","aspect_ratio":"adaptive"}}'
minimax-h3-max-extend

MiniMax H3 Max Extend — 延续现有视频(1.6 至 60 秒),根据对接下来发生内容的描述延长 1–15 秒。Turbo 速度更快,价格为一半;Max 质量更高。结果可以是源视频与延续部分的完整视频,也可以仅包含延续部分。480p / 768p / 1080p / 2K。按新增视频的秒数计费。

变体(addressAs)
参数类型描述
minimax-h3-max-extendflowFlow · 可用
opts 参数
参数类型描述
video_urlreq.text视频(URL)
engineselect模型 · 默认值:turbo · 选项:turbo、max
durationselectSeconds to add · default: 5 · options: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
resolutionselectResolution · default: 768p · options: 480p, 768p, 1080p, 2k
aspect_ratioselect宽高比 · 默认值:auto · 选项:auto、21:9、16:9、4:3、1:1、3:4、9:16
outputselect结果 · 默认值:extended · 选项:extended、continuation
enable_prompt_expansionbool自动扩展提示词 · 默认值:true
seednumberSeed(可选)
  • variants: minimax-h3-max-extend
  • engine: turbo | max
  • duration: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15
  • resolution: 480p | 768p | 1080p | 2k
  • aspect_ratio: auto | 21:9 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16
  • output: extended | continuation

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-max-extend","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_url":"https://example.com/input.mp4","engine":"turbo","duration":"5","resolution":"768p","aspect_ratio":"auto","output":"extended"}}'
minimax-h3-max-insert

MiniMax H3 Max Insert — 在现有视频的 start_time 和 resume_time 之间插入新场景(5–13 秒;均为源视频中的秒数,默认值分别为 2 和 3;start_time 不得早于 1.625 s,且 resume_time 之后必须保留源视频中约 1.5 s 的内容)。颜色会与源视频匹配。提示词用于描述新场景。480p / 768p。按新场景的秒数计费。

变体(addressAs)
参数类型描述
minimax-h3-max-insertflowFlow · 可用
opts 参数
参数类型描述
video_urlreq.text视频(URL)
start_timenumber插入位置,秒 · 默认值:2
resume_timenumber从源视频的此处继续,秒 · 默认值:3
durationselectNew scene length, s · default: 5 · options: 5, 6, 7, 8, 9, 10, 11, 12, 13
resolutionselect分辨率 · 默认值:768p · 选项:480p、768p
color_matchbool使颜色与源视频匹配 · 默认值:true
enable_prompt_expansionbool自动扩展提示词 · 默认值:true
seednumberSeed(可选)
  • variants: minimax-h3-max-insert
  • duration: 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13
  • resolution: 480p | 768p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-max-insert","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_url":"https://example.com/input.mp4","start_time":"2","resume_time":"3","duration":"5","resolution":"768p"}}'
minimax-h3-max-camera

MiniMax H3 Max Camera Controls — 根据一张图生成视频,其中只有摄像机移动:向右、向左或 360° 环绕,推进、拉远、上升或下降。除非提示词另有要求,否则场景保持静止。480p / 768p / 1080p,1–15 秒。按秒计费。

变体(addressAs)
参数类型描述
minimax-h3-max-cameraflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
camera_moveselect摄像机移动 · 默认值:orbit-right · 选项:orbit-right、orbit-left、orbit-360、push-in、pull-out、rise、descend
durationselectDuration, sec · default: 5 · options: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
resolutionselect分辨率 · 默认值:480p · 选项:480p、768p、1080p
prompt_expansion_modeselect提示词扩展 · 默认值:balanced · 选项:balanced、quality、disabled
seednumberSeed(可选)
  • variants: minimax-h3-max-camera
  • camera_move: orbit-right | orbit-left | orbit-360 | push-in | pull-out | rise | descend
  • duration: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15
  • resolution: 480p | 768p | 1080p
  • prompt_expansion_mode: balanced | quality | disabled

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-max-camera","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","camera_move":"orbit-right","duration":"5","resolution":"480p","prompt_expansion_mode":"balanced"}}'
runway-gen-4-5

Runway Gen-4.5 —— 新一代 Runway:根据文本或首帧生成电影感视频,支持 5 或 10 秒片段,以及 6 种宽高比。按片段时长定价。

变体(addressAs)
参数类型描述
runway-gen-4-5canvasCanvas · 可用
opts 参数
参数类型描述
durationselect时长,秒 · 默认值:5 · 选项:5、10
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 9:16, 4:3, 3:4, 1:1, 21:9
image_urltext第一帧(URL,可选)
seednumberSeed(可选)
  • variants: runway-gen-4-5
  • duration: 5 | 10
  • aspect_ratio: 16:9 | 9:16 | 4:3 | 3:4 | 1:1 | 21:9

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"runway-gen-4-5","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","aspect_ratio":"16:9"}}'
p-video-2

P-Video-2(Pruna AI)— 根据文本或首帧快速、低成本地生成有声视频(也可设置尾帧)。支持 720p 或 1080p、2 至 20 秒、7 种宽高比。草稿模式价格减半,适合快速试用。您可以提供现成音轨(audio_url):视频将根据音轨生成,片段时长与音轨一致,并按音轨秒数计费。每个片段按分辨率、模式和时长计费。

变体(addressAs)
参数类型描述
p-video-2canvasCanvas · 可用
opts 参数
参数类型描述
durationselectDuration, sec · default: 5 · options: 2, 3, 5, 8, 10, 15, 20
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
draftbool草稿模式(更便宜,质量较低)· 默认值:false
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 1:1
image_urltext第一帧(URL,可选)
last_frame_urltext末帧(URL,可选)
audio_urltext用于引导视频生成的现成音轨(URL,MP3/WAV/FLAC,可选)
save_audiobool保留片段中的声音 · 默认值:true
seednumberSeed(可选)
  • variants: p-video-2
  • duration: 2 | 3 | 5 | 8 | 10 | 15 | 20
  • resolution: 720p | 1080p
  • aspect_ratio: 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3 | 1:1

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"p-video-2","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"720p","aspect_ratio":"16:9"}}'
p-video-2-pro

P-Video-2 Pro (Pruna AI) — 更高阶的 P-Video-2:根据文本,或根据首帧和末帧生成带声音的视频,时长为 5 至 15 秒,支持 480p 或 768p。提供三种模式:Cost(最便宜)、Speed(快速)和 Quality(最佳质量)。按模式、分辨率和时长以片段计价。

变体(addressAs)
参数类型描述
p-video-2-procanvasCanvas · 可用
opts 参数
参数类型描述
modeselect模式 · 默认值:speed · 选项:cost、speed、quality
durationselectDuration, sec · default: 5 · options: 5, 8, 10, 12, 15
resolutionselect分辨率 · 默认值:768p · 选项:480p、768p
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 1:1
image_urltext第一帧(URL,可选)
last_frame_urltext末帧(URL,可选)
seednumberSeed(可选)
  • variants: p-video-2-pro
  • mode: cost | speed | quality
  • duration: 5 | 8 | 10 | 12 | 15
  • resolution: 480p | 768p
  • aspect_ratio: 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3 | 1:1

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"p-video-2-pro","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"mode":"speed","duration":"5","resolution":"768p","aspect_ratio":"16:9"}}'
p-video-2-pro-reference

P-Video-2 Pro Reference(Pruna AI)— 根据参考素材生成视频:角色或产品照片(最多 9 张)、参考视频(最多 3 个)和音频(最多 3 个),时长 5 至 15 秒,支持 480p 或 768p,以及 Speed 和 Cost 模式。启用 keep_audio 后,您提供的单条音轨将保留为片段音频,声音和台词均保持不变,生成的视频(动作、口型)会与之匹配。未启用 keep_audio 时,音频仅作为模仿样本。按片段计费:输出秒数加上参考素材费用(视频按秒、图像按张、音频按秒计费)。

变体(addressAs)
参数类型描述
p-video-2-pro-referencecanvasCanvas · 可用
opts 参数
参数类型描述
modeselect模式 · 默认值:speed · 选项:speed、cost
durationselectDuration, sec · default: 5 · options: 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
resolutionselect分辨率 · 默认值:768p · 选项:480p、768p
aspect_ratioselectAspect ratio · default: 16:9 · options: adaptive, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 1:1
reference_image_urlsarray参考图像 — 角色、产品、风格(URL,最多 9 张,含首尾帧)
reference_video_urlsarray参考视频(URL,最多 3 个,每个 2–15 秒,总时长 15 秒)
reference_audio_urlsarray音频(URL,最多 3 个,总时长 15 秒)— 作为样本,或在启用 keep_audio 时用作片段配乐
keep_audiobool保留我的音轨作为片段配乐(必须为 1 个文件)· 默认值:false
first_frame_urltext第一帧(URL,可选)
last_frame_urltext末帧(URL,可选)
seednumberSeed(可选)
  • variants: p-video-2-pro-reference
  • mode: speed | cost
  • duration: 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15
  • resolution: 480p | 768p
  • aspect_ratio: adaptive | 16:9 | 9:16 | 4:3 | 3:4 | 3:2 | 2:3 | 1:1

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"p-video-2-pro-reference","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"mode":"speed","duration":"5","resolution":"768p","aspect_ratio":"16:9","reference_image_urls":["https://example.com/input.jpg"],"reference_video_urls":["https://example.com/input.jpg"],"reference_audio_urls":["https://example.com/input.jpg"]}}'
p-video-edit

P-Video Edit (Pruna AI) — 根据文本指令编辑现有视频:替换对象、背景、风格或光照,可使用或不使用参考图像(最多 4 张)。源视频最长 15 秒;结果与源视频等长。草稿模式更便宜。按源视频的每秒计价。

变体(addressAs)
参数类型描述
p-video-editcanvasCanvas · 可用
opts 参数
参数类型描述
video_urlreq.text视频(URL)
input_urlsarray参考图像(URL,最多 4 张)
draftbool草稿模式(更便宜,质量较低)· 默认值:false
seednumberSeed(可选)
  • variants:p-video-edit

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"p-video-edit","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_url":"https://example.com/input.mp4","input_urls":["https://example.com/input.jpg"]}}'
luma-ray-3-2

Luma Ray 3.2 — 新一代 Luma Ray 生成模型(Ray 2 的继任者):根据文本,或根据首帧和末帧,生成电影感无声视频,并支持无缝循环。支持 540p、720p 或 1080p;文本生成支持 5 秒和 10 秒。支持 6 种宽高比。按分辨率和时长以片段计价。

变体(addressAs)
参数类型描述
luma-ray-3-2canvasCanvas · 可用
opts 参数
参数类型描述
resolutionselect分辨率 · 默认值:720p · 选项:540p、720p、1080p
durationselect时长,秒 · 默认值:5 · 选项:5、10
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 9:16, 4:3, 3:4, 1:1, 21:9
image_urltext第一帧(URL,可选)
end_image_urltext末帧(URL,可选)
loopbool无缝循环 · 默认值:false
  • variants: luma-ray-3-2
  • resolution: 540p | 720p | 1080p
  • duration: 5 | 10
  • aspect_ratio: 16:9 | 9:16 | 4:3 | 3:4 | 1:1 | 21:9

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"luma-ray-3-2","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"resolution":"720p","duration":"5","aspect_ratio":"16:9"}}'
veo-3.1

Google Veo 3.1:Lite 使用 Veo 3.1 Lite preview,Fast 使用 Veo 3.1 Fast,Quality 使用 Veo 3.1。所有等级均生成带原生音频的视频;此 API 没有音频开关。支持 720p/1080p/4K、4/6/8 秒;参考模式仅支持 Lite/Fast,并且要求 8 秒。每次任务前请查看当前报价。

变体(addressAs)
参数类型描述
veo-3.1:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
veo-3.1pixelPixel · 可用
veo-3.1:flowflowFlow · 可用
opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16
tierselect质量 · 默认值:fast · 选项:lite、fast、quality
generationTypeselect生成类型 · 默认值:TEXT_2_VIDEO · 选项:TEXT_2_VIDEO、FIRST_AND_LAST_FRAMES_2_VIDEO、REFERENCE_2_VIDEO
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p、4k
durationselect时长,秒 · 默认值:8 · 选项:4、6、8
imageUrlsarrayimageUrls:公开的 HTTP(S) URL(首帧/末帧使用 1–2 个;参考图使用 1–3 个)
  • variants: veo-3.1:white, veo-3.1, veo-3.1:flow
  • 层级:Lite · 720p、Lite · 1080p、Lite · 4k、Fast · 720p、Fast · 1080p、Fast · 4k、Quality · 720p、Quality · 1080p、Quality · 4k(文本 → 视频)、Quality · 4k(帧/参考图)
  • aspect_ratio: 16:9 | 9:16
  • tier: lite | fast | quality
  • generationType: TEXT_2_VIDEO | FIRST_AND_LAST_FRAMES_2_VIDEO | REFERENCE_2_VIDEO
  • resolution: 720p | 1080p | 4k
  • duration: 4 | 6 | 8

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"veo-3.1","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"aspect_ratio":"16:9","tier":"fast","generationType":"TEXT_2_VIDEO","resolution":"720p","duration":"8","imageUrls":["https://example.com/input.jpg"]}}'
kling-v2.6

Kling v2.6 HD:文生视频和图生视频,可选择添加声音。

变体(addressAs)
参数类型描述
kling-v2.6pixelPixel · 可用
kling-v2.6:flowflowFlow · 可用
kling-v2.6:primeprimePrime · 可用
kling-v2.6:whitewhiteWhite · Available
opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
durationselect时长,秒 · 默认值:5 · 选项:5、10
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
resolutionselect分辨率 · 默认值:1080p · 选项:720p、1080p
image_urltext图像(URL,用于 I2V)
audiobool带声音 · 默认值:false
  • variants: kling-v2.6, kling-v2.6:flow, kling-v2.6:prime, kling-v2.6:white
  • 层级:5s · 无音频、10s · 无音频、5s · 有音频、10s · 有音频
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • duration: 5 | 10
  • input: text-to-video | image-to-video (affects price)
  • resolution: 720p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-v2.6","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"aspect_ratio":"16:9","duration":"5","input":"text-to-video","resolution":"1080p"}}'
kling-v3.0

Kling v3.0:最新版本。按分辨率按秒计费,可选择添加声音。模式:std(720p)、pro(1080p)、4K。

变体(addressAs)
参数类型描述
kling-v3.0pixelPixel · 可用
kling-v3.0:primeprimePrime · 可用
opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
durationselect时长,秒 · 默认值:5 · 选项:5、10
modeselect质量模式 · 默认值:pro · 选项:std、pro、4K
multi_shotsselect多镜头 · 默认值:false · 选项:false、true
multi_promptarray镜头(multi_prompt,最多 5 个)
audiobool带声音 · 默认值:false
image_urltext第一帧(URL,可选)
last_frame_urltext末帧(URL,可选)
kling_elementsarray角色元素(最多 3 个)
  • variants: kling-v3.0, kling-v3.0:prime
  • 层级:720p · 无音频、720p · 有音频、1080p · 无音频、1080p · 有音频、4K · 无音频、4K · 有音频
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • duration: 5 | 10
  • mode: std | pro | 4K
  • multi_shots: false | true

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-v3.0","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"aspect_ratio":"16:9","duration":"5","mode":"pro","multi_shots":"false"}}'
kling-v3-omni

通过 Prime 提供的 Kling v3.0 Omni(唯一供应商):支持 T2V 和 I2V,提供 std(720p)或 pro(1080p),无声。按秒计费。Omni 元素/参考功能不可用(Prime 合同无法表达这些功能)。

变体(addressAs)
参数类型描述
kling-v3-omnipixelPixel · 可用
kling-v3-omni:primeprimePrime · 可用
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p、4k
audiobool原生音频 · 默认值:false
durationselect时长,秒 · 默认值:5 · 选项:5、10
image_urltext第一帧(URL,用于 I2V)
last_frame_urltext最后一帧(URL,可选,I2V)
  • variants: kling-v3-omni, kling-v3-omni:prime
  • 层级:720p · 无音频、720p · 有音频、1080p · 无音频、1080p · 有音频、4K · 无音频、4K · 有音频
  • input: text-to-video | image-to-video (affects price)
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • resolution: 720p | 1080p | 4k
  • duration: 5 | 10

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-v3-omni","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","aspect_ratio":"16:9","resolution":"720p","duration":"5"}}'
kling-v3-omni-reference

Kling 3.0 Omni Reference — 根据参考图像(最多 7 张)和/或参考视频生成视频。没有参考视频时才支持原生音频。支持 720p/1080p/4K。根据分辨率、音频以及是否使用参考视频按秒计费。

变体(addressAs)
参数类型描述
kling-v3-omni-referencepixelPixel · 可用
opts 参数
参数类型描述
video_inputselect视频参考 · 默认:without-video · 选项:without-video、with-video
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p、4k
audiobool原生音频 · 默认值:false
durationselectDuration, sec (3–15) · default: 5 · options: 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
image_urltext参考图片(URL)
input_urlsarray参考图像(URL,最多 7 张;与视频一同使用时最多 4 张)
video_urltext参考视频(URL,3–15 s,MP4/MOV)
  • variants: kling-v3-omni-reference
  • 档位:720p · 无音频、720p · 带音频、720p · 带视频、1080p · 无音频、1080p · 带音频、1080p · 带视频、4K · 无音频、4K · 带音频、4K · 带视频
  • video_input: without-video | with-video
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • resolution: 720p | 1080p | 4k
  • duration: 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-v3-omni-reference","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_input":"without-video","aspect_ratio":"16:9","resolution":"720p","duration":"5","input_urls":["https://example.com/input.jpg"]}}'
kling-v3-omni-transformation

Kling 3.0 Omni Transformation — 根据文本描述对源视频(3–15.5 s)进行风格重塑或转换,也可以选择使用参考图像(最多 4 张)。结果无声。支持 720p/1080p/4K。按秒计费;结果会遵循源视频,因此时长选项仅用于估算费用。

变体(addressAs)
参数类型描述
kling-v3-omni-transformationpixelPixel · 可用
opts 参数
参数类型描述
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p、4k
aspect_ratioselect宽高比(仅适用于参考图像)· 默认值:16:9 · 选项:16:9、9:16、1:1
durationselect时长(估算),秒 · 默认值:5 · 选项:5、10、15
video_urlreq.text视频(URL)
image_urltext参考图片(URL)
input_urlsarray参考图像(URL,最多 4 张)
  • variants: kling-v3-omni-transformation
  • tiers: 720p, 1080p, 4K
  • resolution: 720p | 1080p | 4k
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • duration: 5 | 10 | 15

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-v3-omni-transformation","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"resolution":"720p","aspect_ratio":"16:9","duration":"5","video_url":"https://example.com/input.mp4","input_urls":["https://example.com/input.jpg"]}}'
kling-v4

Kling 4.0 — 3 到 30 秒视频,支持原生音频和唇形同步:可从文本、关键帧(首帧、尾帧和最多 8 个中间帧)或参考内容(最多 10 张图片、5 个视频、7 个元素和 3 个声音)生成。支持 720p/1080p。按分辨率以秒计费;输入视频的时长也会计费。

变体(addressAs)
参数类型描述
kling-v4pixelPixel · 可用
opts 参数
参数类型描述
inputselect输入 · 默认:text-to-video · 选项:text-to-video、keyframe-to-video、reference-to-video
aspect_ratioselect宽高比 · 默认:auto · 选项:auto、16:9、9:16、1:1、21:9
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
durationnumber时长(3–30 s)· 默认:5
image_urltext第一帧(URL)
keyframe_urlsarray中间帧(URL,最多 8 个)
last_frame_urltext末帧(URL,可选)
reference_image_urlsarray参考图(URL)
reference_video_urlsarray参考视频(URL)
keep_original_soundbool保留参考视频声音 · 默认:false
elementsarray元素(最多 7 个:名称 + 2–4 张图片或一段 3–8 s 视频)
voicesarray声音(最多 3 个:名称 + audio_url)
  • variants: kling-v4, kling-v4:white, kling-v4:prime
  • input: text-to-video | keyframe-to-video | reference-to-video (affects price)
  • aspect_ratio: auto | 16:9 | 9:16 | 1:1 | 21:9
  • resolution: 720p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-v4","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","aspect_ratio":"auto","resolution":"720p","duration":"5"}}'
kling-v4-flash

Kling 4.0 Flash — 更快、更便宜的 Kling 4.0 档位:支持根据文本、关键帧(首帧、尾帧及最多 8 个中间帧)或参考素材(最多 10 张图片、5 个视频、7 个元素和 3 种声音)生成视频。720p,3–20 秒,原生音频。按秒计费;如有输入视频,其秒数也会计费。

变体(addressAs)
参数类型描述
kling-v4-flashpixelPixel · 可用
opts 参数
参数类型描述
inputselect输入 · 默认:text-to-video · 选项:text-to-video、keyframe-to-video、reference-to-video
aspect_ratioselect宽高比 · 默认:auto · 选项:auto、16:9、9:16、1:1、21:9
resolutionselect分辨率 · 默认值:720p · 可选值:720p
durationnumber时长(3–20 s)· 默认值:5
image_urltext第一帧(URL)
keyframe_urlsarray中间帧(URL,最多 8 个)
last_frame_urltext末帧(URL,可选)
reference_image_urlsarray参考图(URL)
reference_video_urlsarray参考视频(URL)
keep_original_soundbool保留参考视频声音 · 默认:false
elementsarray元素(最多 7 个:名称 + 2–4 张图片或一段 3–8 s 视频)
voicesarray声音(最多 3 个:名称 + audio_url)
  • variants: kling-v4-flash
  • input: text-to-video | keyframe-to-video | reference-to-video (affects price)
  • aspect_ratio: auto | 16:9 | 9:16 | 1:1 | 21:9
  • resolution: 720p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-v4-flash","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","aspect_ratio":"auto","resolution":"720p","duration":"5"}}'
kling-v3-turbo

Kling v3 Turbo:速度快,支持 T2V 和 I2V。按分辨率按秒计费。

opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
durationselect时长,秒 · 默认值:5 · 选项:5、10
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
image_urltext图像(URL,用于 I2V)
  • tiers: 720p, 1080p
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • duration: 5 | 10
  • input: text-to-video | image-to-video (affects price)
  • resolution: 720p | 1080p (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-v3-turbo","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"aspect_ratio":"16:9","duration":"5","input":"text-to-video","resolution":"720p"}}'
kling-v2-5-turbo

Kling v2.5 Turbo:速度快,价格/质量比最佳,支持 T2V 和 I2V。

opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
durationselect时长,秒 · 默认值:5 · 选项:5、10
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
image_urltext图像(URL,用于 I2V)
  • tiers: 5s, 10s
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • duration: 5 | 10
  • input: text-to-video | image-to-video (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-v2-5-turbo","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"aspect_ratio":"16:9","duration":"5","input":"text-to-video"}}'
kling-ai-avatar

Kling AI Avatar 2.0:头像与语音同步。需要 prompt、image_url 和 audio_url,上游服务会拒绝空 prompt。时长由音频文件决定。

opts 参数
参数类型描述
tierselect质量 · 默认值:std · 选项:std、pro
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
  • 层级:Standard 720p、Pro 1080p
  • tier: std | pro (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-ai-avatar","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"tier":"std","image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3"}}'
kling-motion-control

Kling Motion Control:将视频中的动作迁移到图像中的角色上,支持 2.6 和 3.0 版本。需要 image_url 和 video_url。按秒计费。

opts 参数
参数类型描述
versionselect版本 · 默认:3.0 · 选项:2.6、3.0
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
durationselect时长(估计值),秒 · 默认值:5 · 选项:5、10
image_urlreq.text图像(URL)
video_urlreq.text视频(URL)
  • tiers: 2.6 · 720p, 2.6 · 1080p, 3.0 · 720p, 3.0 · 1080p
  • version: 2.6 | 3.0 (affects price)
  • resolution: 720p | 1080p
  • duration: 5 | 10

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kling-motion-control","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"version":"3.0","resolution":"720p","duration":"5","image_url":"https://example.com/input.jpg","video_url":"https://example.com/input.mp4"}}'
seedance-2.0

ByteDance Seedance 2.0,按秒计费。标准/快速 × 分辨率 × 输入。

变体(addressAs)
参数类型描述
seedance-2.0:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
seedance-2.0:pixelpixelPixel · 可用
seedance-2.0:primeprimePrime · 不可用:路由 seedance-2.0:prime 不支持参考图像 (input_urls)。请使用 seedance-2.0 (Pixel)。
opts 参数
参数类型描述
tierselect版本 · 默认:standard · 选项:standard、fast
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
video_inputselect视频参考 · 默认:without-video · 选项:without-video、with-video
resolutionselectResolution · default: 720p · options: 480p, 720p, 1080p, 4K
durationselect时长,秒 · 默认值:5 · 选项:5、10
image_urltext图像(URL,用于 I2V)
last_frame_urltext最后一帧(URL,可选,I2V)
input_urlsarray参考图(1–16 个 URL)
video_urltext参考视频(URL)
reference_audio_urlsarray声音样本:2–30 秒,最多 10 个 URL,每个最多 15 MB,总时长最多 30 秒
web_searchselectWeb 搜索 · 默认值:false · 选项:false、true
audiobool生成声音 · 默认值:false
  • variants: seedance-2.0:white, seedance-2.0:pixel, seedance-2.0:prime
  • 层级:standard · 无视频 · 480p、standard · 无视频 · 720p、standard · 无视频 · 1080p、standard · 无视频 · 4K、standard · 有视频 · 480p、standard · 有视频 · 720p、standard · 有视频 · 1080p、standard · 有视频 · 4K、fast · 无视频 · 480p、fast · 无视频 · 720p、fast · 有视频 · 480p、fast · 有视频 · 720p
  • tier: standard | fast (affects price)
  • input: text-to-video | image-to-video (affects price)
  • video_input: without-video | with-video
  • resolution: 480p | 720p | 1080p | 4K
  • duration: 5 | 10
  • web_search: false | true

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedance-2.0","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"tier":"standard","input":"text-to-video","video_input":"without-video","resolution":"720p","duration":"5","input_urls":["https://example.com/input.jpg"],"reference_audio_urls":["https://example.com/input.jpg"],"web_search":"false"}}'
seedance-2.5

ByteDance Seedance 2.5,根据是否存在视频参考和分辨率按秒计费。支持 T2V、I2V 和视频编辑;编辑要求使用自适应宽高比、时长 -1,且源视频时长为 4–30 秒。

变体(addressAs)
参数类型描述
seedance-2.5:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
seedance-2.5:repcanvasCanvas · 不可用::rep 路由不适用于此模式。请使用 Pixel(不带 :rep)或基础模型。
seedance-2.5:pixelpixelPixel · 可用
seedance-2.5:primeprimePrime · 不可用:路由 seedance-2.5:prime 不支持参考图像 (input_urls)。请使用 seedance-2.5 (Pixel)。
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
video_inputselect视频参考 · 默认:without-video · 选项:without-video、with-video
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1、adaptive
resolutionselect分辨率 · 默认值:720p · 选项:480p、720p、1080p
durationselectDuration, sec · default: 5 · options: 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, -1
image_urltext图像(URL,用于 I2V)
last_frame_urltext最后一帧(URL,可选,I2V)
input_urlsarray参考图(1–16 个 URL)
video_urltext参考视频(URL)
reference_audio_urlsarray声音样本:2–30 秒,最多 10 个 URL,每个最多 15 MB,总时长最多 30 秒
audiobool生成声音 · 默认值:false
  • variants: seedance-2.5:white, seedance-2.5:rep, seedance-2.5:pixel, seedance-2.5:prime
  • 层级:有视频 · 480p、有视频 · 720p、有视频 · 1080p、无视频 · 480p、无视频 · 720p、无视频 · 1080p
  • input: text-to-video | image-to-video (affects price)
  • video_input: without-video | with-video
  • aspect_ratio: 16:9 | 9:16 | 1:1 | adaptive
  • resolution: 480p | 720p | 1080p
  • duration: 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | -1

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedance-2.5","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","video_input":"without-video","aspect_ratio":"16:9","resolution":"720p","duration":"5","input_urls":["https://example.com/input.jpg"],"reference_audio_urls":["https://example.com/input.jpg"]}}'
seedance-2-0-mini

Seedance 2.0 Mini — 经济型 ByteDance 模型,最高支持 720p。按秒计费。

变体(addressAs)
参数类型描述
seedance-2-0-mini:pixelpixelPixel · 可用
seedance-2-0-mini:primeprimePrime · 可用
opts 参数
参数类型描述
resolutionselect分辨率 · 默认值:480p · 选项:480p、720p
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
video_inputselect视频参考 · 默认:without-video · 选项:without-video、with-video
durationselect时长,秒 · 默认值:5 · 选项:5、10
image_urltext图像(URL,用于 I2V)
last_frame_urltext末帧(URL,可选)
audiobool生成声音 · 默认值:false
  • variants: seedance-2-0-mini:pixel, seedance-2-0-mini:prime
  • 层级:无视频 · 480p、无视频 · 720p、有视频 · 480p、有视频 · 720p
  • resolution: 480p | 720p
  • input: text-to-video | image-to-video (affects price)
  • video_input: without-video | with-video
  • duration: 5 | 10

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"seedance-2-0-mini","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"resolution":"480p","input":"text-to-video","video_input":"without-video","duration":"5"}}'
pixverse-v6

PixVerse V6 — 文本/图像/参考/转场/延展。根据质量和声音按秒计价。

变体(addressAs)
参数类型描述
pixverse-v6:pixelpixelPixel · 可用
pixverse-v6:whitewhiteWhite · Available
opts 参数
参数类型描述
modeselect模式 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video、transition、extend
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 4:3, 1:1, 3:4, 9:16, 2:3, 3:2, 21:9
qualityselectQuality · default: 720p · options: 360p, 540p, 720p, 1080p
durationselectDuration, sec · default: 5 · options: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
audiobool生成声音 · 默认值:false
multi_clipbool生成多个片段 · 默认值:false
seednumberSeed(0–2147483647,可选)
image_urltext第一帧图像(URL)
input_urlsarray图像(1–2 个 URL)
last_frame_urltext最后一帧图像(URL)
image_referencesarrayReferences (1–7)
task_idtext要继续执行的已完成任务的 ID
video_urltext用于继续生成的视频 URL
  • variants: pixverse-v6:pixel, pixverse-v6:white
  • 层级:360p · false、360p · true、540p · false、540p · true、720p · false、720p · true、1080p · false、1080p · true
  • mode: text-to-video | image-to-video | reference-to-video | transition | extend (affects price)
  • aspect_ratio: 16:9 | 4:3 | 1:1 | 3:4 | 9:16 | 2:3 | 3:2 | 21:9
  • quality: 360p | 540p | 720p | 1080p
  • duration: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"pixverse-v6","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"mode":"text-to-video","aspect_ratio":"16:9","quality":"720p","duration":"5"}}'
pixverse-v6-max

PixVerse V6 Max — 电影感文本生成视频或首帧视频:支持最长 15 秒、360p–1080p,并可选原生音频。根据分辨率和音频按每秒计费。

变体(addressAs)
参数类型描述
pixverse-v6-max:pixelpixelPixel · 可用
pixverse-v6-max:flowflowFlow · 可用
opts 参数
参数类型描述
modeselect模式 · 默认值:text-to-video · 选项:text-to-video、image-to-video
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 4:3, 1:1, 3:4, 9:16, 2:3, 3:2, 21:9
qualityselectQuality · default: 720p · options: 360p, 540p, 720p, 1080p
durationselectDuration, sec · default: 5 · options: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
audiobool生成声音 · 默认值:false
seednumberSeed(0–2147483647,可选)
image_urltext首帧(URL,用于图像 → 视频)
input_urlsarray首帧(1 个 URL)
  • variants: pixverse-v6-max:pixel, pixverse-v6-max:flow
  • 层级:360p · false、360p · true、540p · false、540p · true、720p · false、720p · true、1080p · false、1080p · true
  • mode: text-to-video | image-to-video (affects price)
  • aspect_ratio: 16:9 | 4:3 | 1:1 | 3:4 | 9:16 | 2:3 | 3:2 | 21:9
  • quality: 360p | 540p | 720p | 1080p
  • duration: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"pixverse-v6-max","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"mode":"text-to-video","aspect_ratio":"16:9","quality":"720p","duration":"5"}}'
wan-3-0

Wan 3.0(Alibaba):支持 480p/720p/1080p 的 T2V 和 I2V。根据分辨率按秒计费。

变体(addressAs)
参数类型描述
wan-3-0canvasCanvas · 可用
wan-3-0:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
modeselect模式 · 默认值:text-to-video · 选项:text-to-video、image-to-video
durationselect时长,秒 · 默认值:5 · 选项:5、10
resolutionselect分辨率 · 默认值:720p · 选项:480p、720p、1080p
aspect_ratioselect宽高比 · 默认值:adaptive · 选项:adaptive、16:9、9:16、1:1、4:3、3:4
image_urltext图像(URL,用于 I2V,即首帧)
negative_prompttext负面提示词(可选)
seednumberSeed(0–2147483647,可选)
prompt_extendbool自动扩展提示词 · 默认值:true
  • variants: wan-3-0, wan-3-0:amber
  • tiers: 480p, 720p, 1080p
  • mode: text-to-video | image-to-video (affects price)
  • duration: 5 | 10
  • resolution: 480p | 720p | 1080p (affects price)
  • aspect_ratio: adaptive | 16:9 | 9:16 | 1:1 | 4:3 | 3:4

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan-3-0","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"mode":"text-to-video","duration":"5","resolution":"720p","aspect_ratio":"adaptive"}}'
wan-3-0-prime

Wan 3.0 Video Prime(Alibaba)— 支持通过文本、首帧(以及末帧)、参考内容(图像、视频、音频),或文件和网页生成视频。时长 2–30 秒,支持 480p/720p/1080p,可带声音或不带声音。按输出视频分辨率以秒计费;模式、宽高比和声音不会改变价格。

变体(addressAs)
参数类型描述
wan-3-0-prime:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
wan-3-0-prime:pixelpixelPixel · 可用
wan-3-0-prime:flowflowFlow · 不可用:wan-3-0-prime:flow 路由不支持文件或链接输入。请使用 wan-3-0-prime:pixel。
opts 参数
参数类型描述
modeselect模式 · 默认值:text-to-video · 选项:text-to-video、first-frame-to-video、first-last-frame-to-video、reference-to-video、file-to-video、link-to-video
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video
durationselectDuration, sec · default: 5 · options: 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30
resolutionselect分辨率 · 默认值:720p · 选项:480p、720p、1080p
aspect_ratioselect宽高比 · 默认值:adaptive · 选项:adaptive、16:9、4:3、1:1、3:4、9:16
audiobool音频 · 默认值:true
first_frame_urltext第一帧(URL)
last_frame_urltext末帧(URL)
reference_image_urlsarray参考图(URL)
reference_video_urlsarray参考视频(URL)
reference_audio_urlsarray参考音频(URLs)
reference_file_urlsarray文件(URL)
reference_link_urlsarray公共页面(URLs)
seednumberSeed
  • variants: wan-3-0-prime:white, wan-3-0-prime:pixel, wan-3-0-prime:flow
  • 层级:480p · 每秒、720p · 每秒、1080p · 每秒
  • mode: text-to-video | first-frame-to-video | first-last-frame-to-video | reference-to-video | file-to-video | link-to-video
  • input: text-to-video | image-to-video | reference-to-video
  • duration: 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30
  • resolution: 480p | 720p | 1080p
  • aspect_ratio: adaptive | 16:9 | 4:3 | 1:1 | 3:4 | 9:16

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan-3-0-prime","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"mode":"text-to-video","input":"text-to-video","duration":"5","resolution":"720p","aspect_ratio":"adaptive","reference_image_urls":["https://example.com/input.jpg"],"reference_video_urls":["https://example.com/input.jpg"],"reference_audio_urls":["https://example.com/input.jpg"],"reference_file_urls":["https://example.com/input.jpg"],"reference_link_urls":["https://example.com/input.jpg"]}}'
flux-3

FLUX 3(Black Forest Labs)— 带同步音频的视频模型:文生视频、图生视频和片段延续。720p/1080p,5–20 秒。按分辨率和模式以秒计费。Draft 是快速预览渲染,仅支持 720p。

变体(addressAs)
参数类型描述
flux-3canvasCanvas · 可用
opts 参数
参数类型描述
modeselect模式 · 默认值:text-to-video · 选项:text-to-video、image-to-video、extend
durationselectDuration, sec · default: 5 · options: 5, 6, 8, 10, 15, 20
qualityselect质量 · 默认值:full · 选项:full、draft
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
aspect_ratioselect宽高比 · 默认值:auto · 选项:auto、21:9、2:1、16:9、4:3、1:1、3:4、9:16
image_urltext图像(URL,用于 I2V)
video_urltext视频(URL,用于延续生成)
generate_audiobool同步音频 · 默认值:true
  • variants: flux-3
  • tiers: 720p, 1080p, Draft 720p, Continuation 720p, Continuation 1080p, Continuation, draft 720p
  • mode: text-to-video | image-to-video | extend (affects price)
  • duration: 5 | 6 | 8 | 10 | 15 | 20
  • quality: full | draft
  • resolution: 720p | 1080p
  • aspect_ratio: auto | 21:9 | 2:1 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"flux-3","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"mode":"text-to-video","duration":"5","quality":"full","resolution":"720p","aspect_ratio":"auto"}}'
wan-2-7

Wan 2.7(Alibaba):T2V、I2V、Video Edit、Reference-to-Video。根据分辨率按秒计费。

变体(addressAs)
参数类型描述
wan-2-7:whitewhiteWhite · Available
wan-2-7:pixelpixelPixel · 可用
opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
durationselect时长,秒 · 默认值:5 · 选项:5、10
modeselect模式 · 默认值:text-to-video · 选项:text-to-video、image-to-video、edit、reference-to-video
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
image_urltext图像(URL,用于 I2V/R2V,即首帧)
last_frame_urltext末帧(URL,用于 I2V)
first_clip_urltext要继续处理的片段(URL,用于 I2V 续接)
video_urltext视频(URL,用于 Edit/R2V)
audio_urltext音频(URL;对于 T2V 为配乐;对于 I2V 为驱动音频)
first_frametext第一帧 R2V(URL)
reference_voicetext参考音频(URL,wav/mp3,1–10s)
negative_prompttext负面提示词(可选)
seedtextSeed(0–2147483647,可选)
prompt_extendbool自动扩展提示词 · 默认值:true
watermarkbool水印“AI generated”· 默认值:false
audio_settingselect音频(用于 Edit) · 默认值:auto · 选项:auto、origin
  • variants: wan-2-7:white, wan-2-7:pixel
  • tiers: T2V · 720p, T2V · 1080p, I2V · 720p, I2V · 1080p, Edit · 720p, Edit · 1080p, R2V · 720p, R2V · 1080p
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • duration: 5 | 10
  • mode: text-to-video | image-to-video | edit | reference-to-video (affects price)
  • resolution: 720p | 1080p (affects price)
  • audio_setting: auto | origin

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan-2-7","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"aspect_ratio":"16:9","duration":"5","mode":"text-to-video","resolution":"720p"}}'
wan-2-6

Wan 2.6:T2V 和 I2V。根据时长和分辨率按片段计费。不支持 aspect_ratio(自动检测)。

变体(addressAs)
参数类型描述
wan-2-6pixelPixel · 可用
wan-2-6:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、flash-image-to-video
durationselect时长,秒 · 默认值:5 · 选项:5、10、15
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
image_urltext图像(URL,用于 I2V)
  • variants: wan-2-6, wan-2-6:amber
  • tiers: 5s · 720p, 5s · 1080p, 10s · 720p, 10s · 1080p, 15s · 720p, 15s · 1080p
  • input: text-to-video | image-to-video | flash-image-to-video (affects price)
  • duration: 5 | 10 | 15
  • resolution: 720p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan-2-6","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","duration":"5","resolution":"720p"}}'
wan-2-5

Wan 2.5:Pixel T2V 和 I2V。根据时长和分辨率按片段计费,两种输入方式的价格相同。

opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
durationselect时长,秒 · 默认值:5 · 选项:5、10
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
image_urltext图像(公共 URL,用于 I2V)
negative_prompttext负面提示词(最多 500 个字符)
enable_prompt_expansionbool提示词扩展 · 默认值:false
seednumberSeed(整数)
nsfw_checkerboolNSFW 检查器 · 默认值:false
  • tiers: 5s · 720p, 5s · 1080p, 10s · 720p, 10s · 1080p
  • input: text-to-video | image-to-video
  • duration: 5 | 10
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • resolution: 720p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan-2-5","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","duration":"5","aspect_ratio":"16:9","resolution":"720p"}}'
wan-2-2

Wan 2.2 A14B Turbo:价格最实惠的 Wan。支持 T2V 和 I2V,分辨率为 480p/580p/720p。

opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
resolutionselect分辨率 · 默认值:720p · 选项:480p、720p
image_urltext图像(URL,用于 I2V)
  • tiers: 480p, 720p
  • aspect_ratio: 16:9 | 9:16
  • input: text-to-video | image-to-video (affects price)
  • resolution: 480p | 720p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan-2-2","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"aspect_ratio":"16:9","input":"text-to-video","resolution":"720p"}}'
wan-speech-to-video

通过 Flow 使用 Wan 2.2 A14B Speech-to-Video Turbo:生成角色说话的视频。需要 prompt、image_url 和 audio_url。始终为 16 fps;num_frames(40–120,且必须是 4 的倍数)设置时长:会渲染 num_frames − 3 帧,80 → 4.81 s,最多 120 → 7.31 s。音频长于视频片段时会被截断。按输出视频的秒数计费。

变体(addressAs)
参数类型描述
wan-speech-to-video:flowflowFlow · 可用
opts 参数
参数类型描述
resolutionselect分辨率 · 默认值:480p · 选项:480p、580p、720p
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
num_framesnumber帧数
frames_per_secondnumberFPS
negative_prompttext负面提示词
num_inference_stepsnumber生成步骤
guidance_scalenumber引导强度
shiftnumberShift
  • variants: wan-speech-to-video:flow
  • resolution: 480p | 580p | 720p (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"wan-speech-to-video","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"resolution":"480p","image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3"}}'
hailuo-2-3

Hailuo 2.3 — 仅支持 I2V Standard 和 I2V Pro。

变体(addressAs)
参数类型描述
hailuo-2-3pixelPixel · 可用
hailuo-2-3:whitewhiteWhite · Available
opts 参数
参数类型描述
tierselect质量 · 默认值:i2v-std · 选项:i2v-std、i2v-pro
durationselect时长,秒 · 默认值:6 · 选项:6、10
resolutionselect分辨率 · 默认值:768p · 选项:768p、1080p
image_urlreq.text图像(URL,必填)
  • variants: hailuo-2-3, hailuo-2-3:white
  • 层级:I2V Std · 6s · 768p、I2V Std · 6s · 1080p、I2V Std · 10s · 768p、I2V Pro · 6s · 768p、I2V Pro · 6s · 1080p、I2V Pro · 10s · 768p
  • tier: i2v-std | i2v-pro (affects price)
  • duration: 6 | 10
  • resolution: 768p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"hailuo-2-3","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"tier":"i2v-std","duration":"6","resolution":"768p","image_url":"https://example.com/input.jpg"}}'
vidu-q3

Vidu Q3——T2V、I2V 和参考图生视频。540p/720p/1080p,时长为 5 或 10 s。按秒计价。

变体(addressAs)
参数类型描述
vidu-q3:whitewhiteWhite · Available
vidu-q3primePrime · 可用
vidu-q3:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video
durationselect时长,秒 · 默认值:5 · 选项:5、10
resolutionselect分辨率 · 默认值:720p · 选项:540p、720p、1080p
image_urltext图像(URL,用于 I2V/R2V)
  • variants: vidu-q3:white, vidu-q3, vidu-q3:amber
  • tiers: 540p, 720p, 1080p
  • input: text-to-video | image-to-video | reference-to-video (affects price)
  • duration: 5 | 10
  • resolution: 540p | 720p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"vidu-q3","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","duration":"5","resolution":"720p"}}'
vidu-q2

由 Prime(唯一提供商)提供的 Vidu Q2 — 支持文生视频、图生视频和参考生成视频。提供 Standard、Pro、Turbo 和 Pro Fast 档位,按秒计费。

变体(addressAs)
参数类型描述
vidu-q2primePrime · 可用
opts 参数
参数类型描述
tierselect模式 · 默认值:t2v · 可选值:t2v, pro, turbo, pro-fast
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video
durationselect时长,秒 · 默认值:5 · 选项:5、10
resolutionselect分辨率 · 默认值:720p · 选项:480p、720p、1080p
image_urltext图像(URL,用于 I2V/R2V)
  • variants: vidu-q2
  • 档位:T2V · 480p, T2V · 720p, T2V · 1080p, Pro · 480p, Pro · 720p, Pro · 1080p, Turbo · 480p, Turbo · 720p, Turbo · 1080p, Pro Fast · 720p, Pro Fast · 1080p
  • tier: t2v | pro | turbo | pro-fast
  • input: text-to-video | image-to-video | reference-to-video
  • duration: 5 | 10
  • resolution: 480p | 720p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"vidu-q2","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"tier":"t2v","input":"text-to-video","duration":"5","resolution":"720p"}}'
vidu-q4

Vidu Q4 — 最新一代 Vidu(Pixel):根据首帧图像生成视频,或根据参考内容生成视频(最多 15 张图像和最多 3 个 MP3 音频片段,每段 3–12s)。原生同步音频,可关闭。时长 3–16s,分辨率 540p/720p/1080p/2K/4K。按分辨率以秒计费。

变体(addressAs)
参数类型描述
vidu-q4:pixelpixelPixel · 可用
vidu-q4:flowflowFlow · 可用
opts 参数
参数类型描述
inputselect输入 · 默认值:image-to-video · 选项:image-to-video、reference-to-video
resolutionselectResolution · default: 720p · options: 540p, 720p, 1080p, 2k, 4k
durationselectDuration, sec · default: 5 · options: 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
audiobool带声音 · 默认值:true
aspect_ratioselectAspect ratio · default: 16:9 · options: 16:9, 9:16, 1:1, 4:3, 3:4
image_urlreq.text图像(URL,用于 I2V)
reference_image_urlsreq.array参考图像(1–15 个 URL,用于 R2V)
reference_audio_urlsarray参考音频(MP3,3–12s,最多 3 个 URL,用于 R2V)
  • variants: vidu-q4:pixel, vidu-q4:flow
  • tiers: 540p, 720p, 1080p, 2K, 4K
  • input: image-to-video | reference-to-video
  • resolution: 540p | 720p | 1080p | 2k | 4k
  • duration: 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16
  • aspect_ratio: 16:9 | 9:16 | 1:1 | 4:3 | 3:4

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"vidu-q4","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"image-to-video","resolution":"720p","duration":"5","image_url":"https://example.com/input.jpg"}}'
happyhorse

HappyHorse(Alibaba)— 版本 1.0 和 1.1。仅支持 T2V、I2V、Reference 和 Video Edit 1.0。按秒计费。Prime 使用最新的 Prime 费率,不支持 Video Edit。

变体(addressAs)
参数类型描述
happyhorse:pixelpixelPixel · 可用
happyhorse:primeprimePrime · 不可用:所选媒体路由不适用于这些参数。
opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
durationselect时长,秒 · 默认值:6 · 选项:6、10
versionselect版本 · 默认:1.0 · 选项:1.0、1.1
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video、video-edit
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
image_urltext图像(URL,用于 I2V/R2V)
reference_imagetext参考图(URL,用于 R2V/编辑)
video_urltext视频(URL,用于 Video Edit)
audio_settingselect音频(用于 Edit) · 默认值:auto · 选项:auto、origin
seedtextSeed(可选)
  • variants: happyhorse:pixel, happyhorse:prime
  • tiers: 1.0 · 720p, 1.0 · 1080p, 1.1 · 720p, 1.1 · 1080p
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • duration: 6 | 10
  • version: 1.0 | 1.1 (affects price)
  • input: text-to-video | image-to-video | reference-to-video | video-edit (affects price)
  • resolution: 720p | 1080p
  • audio_setting: auto | origin

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"happyhorse","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"aspect_ratio":"16:9","duration":"6","version":"1.0","input":"text-to-video","resolution":"720p"}}'
grok-imagine-video

Grok Imagine Video (xAI) — 文生视频和图生视频。版本包括 1.0(文生视频/图生视频使用独立的 slug)和 1.5 预览版。按分辨率按秒计费。Spicy 模式不适用于图生视频(外部图像)。版本 1.5:支持首帧和尾帧(first_frame_url、last_frame_url)、循环帧 loop_url 以及最多 4 个关键帧({url, timestamp_s},时间戳为大于 0 且小于视频时长的整数秒);这些功能不影响价格。

变体(addressAs)
参数类型描述
grok-imagine-videopixelPixel · 可用
grok-imagine-video:whitewhite白色 · 不可用:所选媒体路由不支持这些参数。
opts 参数
参数类型描述
versionselect版本 · 默认:1.0 · 选项:1.0、1.5
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
modeselect运动模式 · 默认值:normal · 选项:normal、fun、spicy
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
resolutionselect分辨率 · 默认值:720p · 选项:480p、720p、1080p
durationselect时长,秒 · 默认值:6 · 选项:6、10
image_urltext图像(URL,用于 I2V)
first_frame_urltext第一帧(URL,可选)
last_frame_urltext末帧(URL,可选)
loop_urltext循环帧 — 视频片段的起始和结束帧(URL,可选)
keyframesarray关键帧(最多 4 个:url + timestamp_s,整数秒)
  • 变体:grok-imagine-video, grok-imagine-video:white
  • tiers: 1.0 · 480p, 1.0 · 720p, 1.0 · 1080p, 1.5 · 480p, 1.5 · 720p
  • version: 1.0 | 1.5 (affects price)
  • input: text-to-video | image-to-video (affects price)
  • mode: normal | fun | spicy
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • resolution: 480p | 720p | 1080p (affects price)
  • duration: 6 | 10

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-imagine-video","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"version":"1.0","input":"text-to-video","mode":"normal","aspect_ratio":"16:9","resolution":"720p","duration":"6"}}'
grok-imagine-upscale

Grok Imagine Upscale — 将 Pixel 生成的视频放大到 1080p。需要成功的 Pixel 视频生成任务的 task_id(不支持外部视频)。

opts 参数
参数类型描述
task_idreq.text成功的 Pixel 视频任务的 ID(源任务)
resolutionselect源视频分辨率(用于计价)· 默认值:720p · 可选值:480p、720p
  • tiers: 480P → 1080P, 720P → 1080P
  • resolution: 480p | 720p (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-imagine-upscale","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"resolution":"720p"}}'
grok-imagine-extend

Grok Imagine Video Extend — 继续在 Pixel 上生成的视频片段(不接受外部视频)。按片段计费,取决于两个维度:延长时长(6 或 10 s)以及源片段的分辨率。

opts 参数
参数类型描述
task_idreq.text成功的 Pixel 视频任务的 ID(源任务)
extend_timesselect扩展时长,秒 · 默认值:6 · 选项:6、10
source_resolutionselect源视频分辨率(用于计费) · 默认值:720p · 可选值:480p、720p
extend_atnumber从第二个继续(≥2)
  • 层级:6 s · 480p、6 s · 720p、10 s · 480p、10 s · 720p
  • extend_times: 6 | 10
  • source_resolution: 480p | 720p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-imagine-extend","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"extend_times":"6","source_resolution":"720p"}}'
minimax-h3-max

Flow Research H3 Max — 一个独立于 MiniMax H3、经过后训练的模型。支持 480P/768P 的文生视频和图生视频,时长为 5–15 秒。公布的目录价格:480P 为 $0.05/s,768P 为 $0.08/s。

变体(addressAs)
参数类型描述
minimax-h3-maxflowFlow · 可用
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
resolutionselect分辨率 · 默认值:768p · 选项:480p、768p
durationselectDuration, sec · default: 5 · options: 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
aspect_ratioselectAspect ratio · default: 16:9 · options: 21:9, 16:9, 4:3, 1:1, 3:4, 9:16
image_urltext图像(URL,用于 I2V)
last_frame_urltext末帧(URL,用于 I2V)
target_audio_urltext现成音轨 — 替换片段配乐(音频 URL,至少 2 秒,最大 15 MB,可选)
prompt_expansion_modeselect提示词扩展 · 默认值:balanced · 选项:balanced、quality
seednumberSeed
enable_safety_checkerbool安全检查器 · 默认值:true
  • variants: minimax-h3-max
  • input: text-to-video | image-to-video
  • resolution: 480p | 768p
  • duration: 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15
  • aspect_ratio: 21:9 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16
  • prompt_expansion_mode: balanced | quality

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-max","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","resolution":"768p","duration":"5","aspect_ratio":"16:9","prompt_expansion_mode":"balanced"}}'
minimax-h3-max-turbo

通过 Flow 使用 MiniMax H3 Max Turbo:文本转视频和图像转视频。按 480p 或 768p 的输出秒数计费;促销价格截至 September 7, 2026。

变体(addressAs)
参数类型描述
minimax-h3-max-turboflowFlow · 可用
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video
resolutionselect分辨率 · 默认值:480p · 选项:480p、768p
durationselectDuration, sec · default: 5 · options: 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
aspect_ratioselectAspect ratio · default: 16:9 · options: 21:9, 16:9, 4:3, 1:1, 3:4, 9:16
image_urltext图像(URL,用于 I2V)
last_frame_urltext末帧(URL,用于 I2V)
target_audio_urltext现成音轨 — 替换片段配乐(音频 URL,至少 2 秒,最大 15 MB,可选)
prompt_expansion_modeselect提示词扩展 · 默认值:balanced · 选项:balanced、quality
seednumberSeed
enable_safety_checkerbool安全检查器 · 默认值:true
  • variants: minimax-h3-max-turbo
  • input: text-to-video | image-to-video
  • resolution: 480p | 768p
  • duration: 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15
  • aspect_ratio: 21:9 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16
  • prompt_expansion_mode: balanced | quality

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-max-turbo","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","resolution":"480p","duration":"5","aspect_ratio":"16:9","prompt_expansion_mode":"balanced"}}'
gemini-omni-flash-1-1-flow

通过 Flow 提供的 Gemini Omni Flash v1.1,支持文生视频、图生视频、参考图生视频和编辑。按 360p/720p/1080p/4K 的输出秒数计费。

变体(addressAs)
参数类型描述
gemini-omni-flash-1-1-flowflowFlow · 可用
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video、edit
resolutionselectResolution · default: 720p · options: 360p, 720p, 1080p, 4k
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16
image_urltextImage/reference (URL)
video_urltext视频(URL,用于编辑)
end_image_urltext末帧(URL)
image_urlsarray参考图(URL)
reference_video_urlsarray参考视频(URL)
source_duration_secondsnumber源视频时长,秒(用于编辑)
durationselectDuration, sec · default: 5 · options: 5, 6, 7, 8, 9, 10
  • variants: gemini-omni-flash-1-1-flow
  • input: text-to-video | image-to-video | reference-to-video | edit
  • resolution: 360p | 720p | 1080p | 4k
  • aspect_ratio: 16:9 | 9:16
  • duration: 5 | 6 | 7 | 8 | 9 | 10

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-omni-flash-1-1-flow","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","resolution":"720p","aspect_ratio":"16:9","duration":"5"}}'
minimax-h3

MiniMax H3 (Hailuo 03):T2V、I2V 和参考图转视频。按分辨率以输出秒数计费。

变体(addressAs)
参数类型描述
minimax-h3pixelPixel · 可用
minimax-h3:whitewhiteWhite · Available
minimax-h3:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video
aspect_ratioselectAspect ratio · default: 16:9 · options: 21:9, 16:9, 4:3, 1:1, 3:4, 9:16
resolutionselect分辨率 · 默认值:768p · 选项:768p、2k
durationselectDuration, sec · default: 6 · options: 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
first_frame_urltext第一帧(URL,用于 I2V)
last_frame_urltext最后一帧(URL,用于 I2V)
reference_image_urlsarray参考图(1–9 个 URL,用于 R2V)
reference_video_urlsarray参考视频(最多 3 个 URL,用于 R2V)
reference_audio_urlsarray参考音频(最多 3 个 URLs,用于 R2V)
input_video_durationnumber输入视频时长,秒(用于 R2V)
  • variants: minimax-h3, minimax-h3:white, minimax-h3:amber
  • tiers: 768p, 2K
  • input: text-to-video | image-to-video | reference-to-video (affects price)
  • aspect_ratio: 21:9 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16
  • resolution: 768p | 2k
  • duration: 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","aspect_ratio":"16:9","resolution":"768p","duration":"6"}}'
minimax-h3-lite

MiniMax H3 Lite — 更轻量的 MiniMax H3:支持文生视频、首尾帧视频生成和参考素材生成(最多 9 张图片、3 个视频和 3 段音频)。时长 4–15 秒,分辨率为 480P 或 768P。根据分辨率按秒计费。

变体(addressAs)
参数类型描述
minimax-h3-litecanvasCanvas · 可用
minimax-h3-lite:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video
aspect_ratioselectAspect ratio · default: 16:9 · options: 21:9, 16:9, 4:3, 1:1, 3:4, 9:16
resolutionselect分辨率 · 默认值:768p · 选项:480p、768p
durationselectDuration, sec · default: 5 · options: 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
first_frame_urltext第一帧(URL,用于 I2V)
last_frame_urltext最后一帧(URL,用于 I2V)
reference_image_urlsarray参考图(1–9 个 URL,用于 R2V)
reference_video_urlsarray参考视频(最多 3 个 URL,用于 R2V)
reference_audio_urlsarray参考音频(最多 3 个 URLs,用于 R2V)
  • variants: minimax-h3-lite, minimax-h3-lite:amber
  • tiers: 480p, 768p
  • input: text-to-video | image-to-video | reference-to-video (affects price)
  • aspect_ratio: 21:9 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16
  • resolution: 480p | 768p (affects price)
  • duration: 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-lite","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","aspect_ratio":"16:9","resolution":"768p","duration":"5"}}'
minimax-h3-highspeed

MiniMax H3 Highspeed — 速度更快的 MiniMax H3:支持文生视频、首尾帧视频生成和参考素材生成(最多 9 张图片、3 个视频和 3 段音频)。时长 4–15 秒,分辨率为 768P。按秒计费。

变体(addressAs)
参数类型描述
minimax-h3-highspeedcanvasCanvas · 可用
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video
aspect_ratioselectAspect ratio · default: 16:9 · options: 21:9, 16:9, 4:3, 1:1, 3:4, 9:16
resolutionselect分辨率 · 默认值:768p · 可选值:768p
durationselectDuration, sec · default: 5 · options: 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
first_frame_urltext第一帧(URL,用于 I2V)
last_frame_urltext最后一帧(URL,用于 I2V)
reference_image_urlsarray参考图(1–9 个 URL,用于 R2V)
reference_video_urlsarray参考视频(最多 3 个 URL,用于 R2V)
reference_audio_urlsarray参考音频(最多 3 个 URLs,用于 R2V)
  • variants: minimax-h3-highspeed
  • tiers: 768p
  • input: text-to-video | image-to-video | reference-to-video (affects price)
  • aspect_ratio: 21:9 | 16:9 | 4:3 | 1:1 | 3:4 | 9:16
  • resolution: 768p(影响价格)
  • duration: 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-highspeed","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","aspect_ratio":"16:9","resolution":"768p","duration":"5"}}'
gemini-omni-video

Gemini Omni Video,多模态生成。声音使用来自 POST /api/media/gemini-omni-audio 的最多 3 个 audio_ids。这些是预设语音,不是根据录音进行的声音克隆,也不是对上传音轨进行的口型同步。若要使用克隆声音生成精确脚本,请使用 voice-clone-tts 和 omnihuman-1-5。

opts 参数
参数类型描述
inputselect视频输入 · 默认:no-video · 选项:no-video、video
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16
durationselect时长,秒 · 默认值:8 · 选项:4、6、8、10
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p、4k
image_urltext图像(URL,可选)
audio_idsarray最多 3 个 audio_ids — POST /api/media/gemini-omni-audio
video_listarray参考视频(遵循 Pixel schema 的 video_list 对象)
seednumberSeed(可选)
  • 层级:无视频 · 4 sec · 720p、无视频 · 4 sec · 1080p、无视频 · 4 sec · 4K、无视频 · 6 sec · 720p、无视频 · 6 sec · 1080p、无视频 · 6 sec · 4K、无视频 · 8 sec · 720p、无视频 · 8 sec · 1080p、无视频 · 8 sec · 4K、无视频 · 10 sec · 720p、无视频 · 10 sec · 1080p、无视频 · 10 sec · 4K、有视频 · 720p、有视频 · 1080p、有视频 · 4K
  • input: no-video | video
  • aspect_ratio: 16:9 | 9:16
  • duration: 4 | 6 | 8 | 10
  • resolution: 720p | 1080p | 4k

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-omni-video","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"no-video","aspect_ratio":"16:9","duration":"8","resolution":"720p"}}'
gemini-omni-flash-1-1

Gemini Omni 1.1 Flash,多模态生成。声音使用来自 POST /api/media/gemini-omni-audio 的最多 3 个 audio_ids。这些是预设语音,不是根据录音进行的声音克隆,也不是对上传音轨进行口型同步。请勿将 audio_ids 与 first_frame_url 结合使用。若要使用克隆声音生成精确脚本,请使用 voice-clone-tts 和 omnihuman-1-5。

opts 参数
参数类型描述
inputselect视频输入 · 默认:no-video · 选项:no-video、video
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16
durationselect时长,秒 · 默认值:8 · 选项:4、6、8、10
resolutionselectResolution · default: 720p · options: 360p, 720p, 1080p, 4k
image_urltext图像(URL,可选)
audio_idsarray最多 3 个 audio_ids — POST /api/media/gemini-omni-audio
video_listarray视频输入(符合 Pixel 架构的 video_list 对象)
seednumberSeed(可选)
  • 层级:无视频 · 4 sec · 360p、无视频 · 4 sec · 720p、无视频 · 4 sec · 1080p、无视频 · 4 sec · 4K、无视频 · 6 sec · 360p、无视频 · 6 sec · 720p、无视频 · 6 sec · 1080p、无视频 · 6 sec · 4K、无视频 · 8 sec · 360p、无视频 · 8 sec · 720p、无视频 · 8 sec · 1080p、无视频 · 8 sec · 4K、无视频 · 10 sec · 360p、无视频 · 10 sec · 720p、无视频 · 10 sec · 1080p、无视频 · 10 sec · 4K、有视频 · 360p、有视频 · 720p、有视频 · 1080p、有视频 · 4K
  • input: no-video | video
  • aspect_ratio: 16:9 | 9:16
  • duration: 4 | 6 | 8 | 10
  • resolution: 360p | 720p | 1080p | 4k

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-omni-flash-1-1","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"no-video","aspect_ratio":"16:9","duration":"8","resolution":"720p"}}'
gemini-omni-flash-1-2

Gemini Omni 1.2 Flash (Google) — 根据文本、首帧(及尾帧)或参考素材生成视频,最长 15 秒。支持 360p/720p/1080p/4K,宽高比为 16:9 或 9:16。原生音频:自动生成语音、音乐和音效;不接受音频输入。最多支持 2 个角色(使用来自 POST /api/media/gemini-omni-character 的 character_ids,不能与首帧/尾帧同时使用),价格不变。如需延长已完成的片段,请使用 gemini-omni-flash-1-2-extend。按分辨率和输出秒数计费。

变体(addressAs)
参数类型描述
gemini-omni-flash-1-2pixelPixel · 可用
gemini-omni-flash-1-2:whitewhiteWhite · Available
opts 参数
参数类型描述
inputselect输入 · 默认值:text-to-video · 选项:text-to-video、image-to-video、reference-to-video
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16
durationselectDuration, sec (up to 10 with a last frame) · default: 5 · options: 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
resolutionselectResolution · default: 720p · options: 360p, 720p, 1080p, 4k
image_urltext图像(URL,用于 I2V)
last_frame_urltext末帧(URL,用于 I2V)
image_urlsarray参考图(URL)
character_idsarray角色 — 最多 2 个来自 POST /api/media/gemini-omni-character 的 id
  • variants: gemini-omni-flash-1-2, gemini-omni-flash-1-2:white
  • 层级:360p · 每秒、720p · 每秒、1080p · 每秒、4K · 每秒
  • input: text-to-video | image-to-video | reference-to-video
  • aspect_ratio: 16:9 | 9:16
  • duration: 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15
  • resolution: 360p | 720p | 1080p | 4k

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-omni-flash-1-2","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"input":"text-to-video","aspect_ratio":"16:9","duration":"5","resolution":"720p"}}'
gemini-omni-flash-1-2-extend

Gemini Omni 1.2 Flash Extend — 根据后续情节描述,将已完成的 Gemini Omni 1.2 Flash 片段延长 3–10 秒。源片段通过 jobId 指定,须为最长 15 秒的 360p 或 720p 片段(只要源片段不超过 15 秒,即可连续延长)。结果为完整片段(源片段 + 续接部分),分辨率和宽高比与源片段一致。新部分最多支持 3 张参考图像。按源片段分辨率和新增秒数计费。

opts 参数
参数类型描述
source_task_idreq.text源 Gemini Omni 1.2 Flash 片段的 jobId
durationselectSeconds to add · default: 5 · options: 3, 4, 5, 6, 7, 8, 9, 10
resolutionselect源片段分辨率(用于确定费率)· 默认:720p · 选项:360p, 720p
image_urlsarray续接部分的参考图像(URL)
  • 档位:360p · 每新增一秒,720p · 每新增一秒
  • duration: 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10
  • resolution: 360p | 720p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-omni-flash-1-2-extend","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"duration":"5","resolution":"720p","image_urls":["https://example.com/input.jpg"]}}'
runway-gen4

Runway Gen4 —— 电影级画质。支持 T2V 和 I2V、720p/1080p。

opts 参数
参数类型描述
aspect_ratioselect宽高比 · 默认值:16:9 · 选项:16:9、9:16、1:1
durationselect时长,秒 · 默认值:5 · 选项:5、10
qualityselect分辨率 · 默认值:720p · 选项:720p、1080p
image_urltext图像(URL,用于 I2V,可选)
  • tiers: 5s · 720p, 5s · 1080p, 10s · 720p
  • aspect_ratio: 16:9 | 9:16 | 1:1
  • duration: 5 | 10
  • quality: 720p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"runway-gen4","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"aspect_ratio":"16:9","duration":"5","quality":"720p"}}'
omnihuman-1-5

OmniHuman 1.5 — 根据照片和语音让人物动起来。需要 image_url 和 audio_url。按秒计价。上一代版本:我们推荐使用 omnihuman-v2——更新且更便宜。

opts 参数
参数类型描述
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
output_resolutionselect分辨率 · 默认值:1080 · 选项:720、1080
pe_fast_modebool快速模式 · 默认值:false
seednumberSeed (-1 = random)
mask_urltext主体蒙版(URL,可选)
  • output_resolution: 720 | 1080

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"omnihuman-1-5","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3","output_resolution":"1080"}}'
omnihuman-2-0

OmniHuman 2.0 — 从一张照片和一段音轨生成会说话的人物:嘴唇、面部表情、手势和身体动作会跟随您的音频。需要 image_url 和 audio_url(wav 或 mp3,4 秒至 5 分钟);视频片段时长与音频相同。可选的 prompt 可引导场景和镜头。720p。按输出视频的每秒计费。

opts 参数
参数类型描述
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
enable_watermarkbool水印 · 默认值:false

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"omnihuman-2-0","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3"}}'
volcengine-lipsync

Volcengine Lip Sync:唇形同步。需要 mode、video_url 和 audio_url。按秒计费。

opts 参数
参数类型描述
modeselect模式 · 默认值:lite · 选项:lite、basic
video_urlreq.text视频(URL)
audio_urlreq.text音频(URL)
separate_vocalbool分离人声(抑制背景噪音) · 默认值:false
open_scenedetbool场景分段和说话人检测 · 默认值:false
align_audiobool如果音频更长,则循环播放视频 · 默认值:true
  • mode: lite | basic

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"volcengine-lipsync","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"mode":"lite","video_url":"https://example.com/input.mp4","audio_url":"https://example.com/input.mp3"}}'
topaz-video-upscale

Topaz Video Upscale — 使用 Topaz Video AI 模型(Proteus、Artemis、Gaia、Nyx)提升现有视频的分辨率:细节更清晰、噪点更少、压缩伪影更少。发送 video_url(MP4/MOV,最长 60 秒)以及 1 到 4 之间的 upscale_factor——例如,×3 可将 360p 转为 1080p,×2 可将 1080p 转为 4K。可选的 target_fps(最高 60)可启用帧插值。按源视频的输出分辨率计费:最高 720p、最高 1080p,或高于 1080p(4K);输出帧率高于 30 fps 时,费用加倍。任务开始前,服务器会自行测量帧尺寸、帧率和时长。

变体(addressAs)
参数类型描述
topaz-video-upscaleflowFlow · 可用
opts 参数
参数类型描述
video_urlreq.text视频(URL)
upscale_factorselectScale · default: 2 · options: 1, 2, 3, 4
modelselectTopaz 模型 · 默认值:Proteus · 选项:Proteus、Artemis HQ、Artemis MQ、Artemis LQ、Gaia HQ、Gaia CG、Nyx、Nyx Fast、Nyx XL、Nyx HF
target_fpsselect输出帧率 · 默认值:source · 选项:source、24、25、30、48、50、60
  • 变体:topaz-video-upscale
  • 层级:输出最高 720p,最高 30 fps;输出最高 1080p,最高 30 fps;输出高于 1080p(4K),最高 30 fps
  • upscale_factor: 1 | 2 | 3 | 4
  • model: Proteus | Artemis HQ | Artemis MQ | Artemis LQ | Gaia HQ | Gaia CG | Nyx | Nyx Fast | Nyx XL | Nyx HF
  • target_fps: source | 24 | 25 | 30 | 48 | 50 | 60

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"topaz-video-upscale","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_url":"https://example.com/input.mp4","upscale_factor":"2","model":"Proteus","target_fps":"source"}}'
video-upscale

Video Upscale — 将现有视频提升至 1080P、2K 或 4K,同时保留宽高比、帧率、帧数和音频。请提供公开可访问的 video_url(MP4/MOV/WEBM,最大 50 MB,最长 60 秒)。分辨率指定输出边界框:画面会按比例缩放至 1920×1080 / 2560×1440 / 3840×2160(竖屏为 1080×1920 / 1440×2560 / 2160×3840)范围内,绝不裁剪;尺寸更大的源视频会缩小至该边界框内。输出格式为 MP4 H.264 + AAC。按输入视频时长逐秒计费,单价取决于输出分辨率:1080P、2K(默认)或 4K;服务器会在任务开始前测量视频片段的时长。

opts 参数
参数类型描述
video_urlreq.text视频(URL)
resolutionselect输出分辨率 · 默认值:2K · 可选值:1080P, 2K, 4K
  • tiers: 1080P · 1920×1080, 2K · 2560×1440, 4K · 3840×2160
  • resolution: 1080P | 2K | 4K

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"video-upscale","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_url":"https://example.com/input.mp4","resolution":"2K"}}'
omnihuman-v2

OmniHuman v2 — 新一代 OmniHuman(2026 年 10 月)以及推荐版本:根据照片和语音轨道生成逼真的人物动画——嘴唇、面部表情、手势和身体动作都会跟随你的音频。需要 image_url 和 audio_url(wav 或 mp3,4 秒至 5 分钟);片段时长与音频相同。可选的提示词(使用英文)可控制场景和镜头。比 OmniHuman 1.5 更便宜。按成片视频的每秒计价。

变体(addressAs)
参数类型描述
omnihuman-v2:flowflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
  • variants: omnihuman-v2:flow

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"omnihuman-v2","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3"}}'
omnihuman-v1.5

OmniHuman v1.5 — 通过 Flow 根据图像和音频让人物动起来。时长取自提供商的结果。上一代版本:我们推荐使用 omnihuman-v2——更新且更便宜。

变体(addressAs)
参数类型描述
omnihuman-v1.5:flowflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
mask_urltext蒙版(URL,可选)
resolutionselect分辨率 · 默认值:1080p · 选项:720p、1080p
turbo_modeboolTurbo 模式 · 默认值:false
  • variants: omnihuman-v1.5:flow
  • resolution: 720p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"omnihuman-v1.5","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3","resolution":"1080p"}}'
lipsync-2-pro

Lipsync-2-pro:将现有视频中的嘴唇动作重新同步到新音频(替换片段中的一行)。需要 video_url 和 audio_url。使用 sync_mode=cut_off(默认)时,输出长度为 min(video, audio),因此相同长度会保持原时长。按输出视频秒数计费。

变体(addressAs)
参数类型描述
lipsync-2-pro:flowflowFlow · 可用
opts 参数
参数类型描述
video_urlreq.text视频(URL)
audio_urlreq.text音频(URL)
sync_modeselect长度不同时 · 默认值:cut_off · 选项:cut_off、loop、bounce、remap、silence
  • variants: lipsync-2-pro:flow
  • sync_mode: cut_off | loop | bounce | remap | silence

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"lipsync-2-pro","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_url":"https://example.com/input.mp4","audio_url":"https://example.com/input.mp3","sync_mode":"cut_off"}}'
sync-3

Sync-3 — 将现有视频与新音频进行新一代唇形同步;在人像侧面、遮挡和特写场景下效果更好。需要 video_url 和 audio_url。使用 sync_mode=cut_off(默认)时,输出时长为 min(video, audio)。按输出视频秒数计费。

变体(addressAs)
参数类型描述
sync-3:flowflowFlow · 可用
opts 参数
参数类型描述
video_urlreq.text视频(URL)
audio_urlreq.text音频(URL)
sync_modeselect长度不同时 · 默认值:cut_off · 选项:cut_off、loop、bounce、remap、silence
  • variants: sync-3:flow
  • sync_mode: cut_off | loop | bounce | remap | silence

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"sync-3","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_url":"https://example.com/input.mp4","audio_url":"https://example.com/input.mp3","sync_mode":"cut_off"}}'
react-1

React-1 — 修改片段中的动作以匹配音频:嘴唇、面部表情(情绪),以及在 model_mode=head 时的头部动作。视频和音频最长 15 s,且只能有一名可见说话者。使用 sync_mode=cut_off(默认值)时,输出长度为 min(video, audio)。按输出视频秒数计费。

变体(addressAs)
参数类型描述
react-1:flowflowFlow · 可用
opts 参数
参数类型描述
video_urlreq.text视频(URL)
audio_urlreq.text音频(URL)
emotionselect情绪 · 默认值:neutral · 选项:neutral、happy、sad、angry、disgusted、surprised
model_modeselect要更改的内容 · 默认值:face · 选项:lips、face、head
temperaturenumberExpressiveness (0–1)
sync_modeselect长度不同时 · 默认值:cut_off · 选项:cut_off、loop、bounce、remap、silence
  • variants: react-1:flow
  • emotion: neutral | happy | sad | angry | disgusted | surprised
  • model_mode: lips | face | head
  • sync_mode: cut_off | loop | bounce | remap | silence

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"react-1","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"video_url":"https://example.com/input.mp4","audio_url":"https://example.com/input.mp3","emotion":"neutral","model_mode":"face","sync_mode":"cut_off"}}'
minimax-h3-max-lipsync

MiniMax H3 Max Lip Sync:使用一张照片和一条音轨生成会说话的视频:嘴唇、表情和头部动作会跟随你的配音,包括俄语。需要 image_url(宽高比 0.4–2.5)和 audio_url(5 秒至 15 分钟);片段长度与音频相同。480p / 768p / 1080p / 2K;启用 enable_transcription 可转录语音,使嘴唇同步更加精准。按输出视频秒数计费,超过 15 s 的片段按 ×1.2 计费。

变体(addressAs)
参数类型描述
minimax-h3-max-lipsync:flowflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
resolutionselectResolution · default: 768p · options: 480p, 768p, 1080p, 2k
enable_transcriptionbool转录语音以实现更紧密的唇形同步 · 默认值:false
seednumberSeed
  • variants: minimax-h3-max-lipsync:flow
  • resolution: 480p | 768p | 1080p | 2k

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-max-lipsync","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3","resolution":"768p"}}'
sync-3-avatar

Sync-3 Avatar — 从一张照片和一条音轨生成会说话的角色(与片段唇形同步使用的 sync-3 相同):唇形精准,也适用于插画。需要 image_url 和 audio_url;片段时长与音频相同(最长 5 分钟)。按输出视频秒数计费。

变体(addressAs)
参数类型描述
sync-3-avatar:flowflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
  • variants: sync-3-avatar:flow

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"sync-3-avatar","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3"}}'
heygen-avatar-4

HeyGen Avatar IV — 使用一张照片和你的配音生成会说话的头像:嘴唇、表情和手势。需要提供 image_url,且其中应包含清晰可见的面部,还需要 audio_url;片段时长与音频相同(最长 5 分钟)。talking_style:stable(平静)或 expressive(富有表现力)。支持 360p–1080p。按输出视频的秒数计费。

变体(addressAs)
参数类型描述
heygen-avatar-4:flowflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
resolutionselectResolution · default: 720p · options: 360p, 480p, 540p, 720p, 1080p
aspect_ratioselect宽高比 · 默认值:auto · 选项:auto、16:9、9:16、4:5、5:4、1:1
talking_styleselect说话风格 · 默认值:stable · 选项:stable、expressive
  • variants: heygen-avatar-4:flow
  • resolution: 360p | 480p | 540p | 720p | 1080p
  • aspect_ratio: auto | 16:9 | 9:16 | 4:5 | 5:4 | 1:1
  • talking_style: stable | expressive

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"heygen-avatar-4","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3","resolution":"720p","aspect_ratio":"auto","talking_style":"stable"}}'
heygen-avatar-v

HeyGen Avatar V — 最新的口播数字人引擎:使用现成的 HeyGen 数字人播报您的配音,动作和唇形同步比 Avatar IV 更自然。需要提供 avatar(现成的 HeyGen 数字人名称,例如“Ann Doctor Standing”)和 audio_url;视频片段时长与音频相同(最长 5 分钟)。720p / 1080p / 4K。按输出视频的秒数计费。

变体(addressAs)
参数类型描述
heygen-avatar-v:flowflowFlow · 可用
opts 参数
参数类型描述
avatarreq.textHeyGen 数字人(现成的数字人名称)
audio_urlreq.text音频(URL)
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p、4k
aspect_ratioselect宽高比 · 默认值:auto · 选项:auto、16:9、9:16、4:5、5:4、1:1
  • variants: heygen-avatar-v:flow
  • resolution: 720p | 1080p | 4k
  • aspect_ratio: auto | 16:9 | 9:16 | 4:5 | 5:4 | 1:1

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"heygen-avatar-v","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"avatar":"Ann Doctor Standing","audio_url":"https://example.com/input.mp3","resolution":"720p","aspect_ratio":"auto"}}'
veed-fabric-1.0

VEED Fabric 1.0 — 根据一张照片和一段音轨生成会说话的视频:嘴唇和表情会跟随你的配音。需要 image_url 和 audio_url(每次请求最长 60 秒);片段长度与音频相同。480p / 720p。按输出视频秒数计费。

变体(addressAs)
参数类型描述
veed-fabric-1.0:flowflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
audio_urlreq.text音频(URL)
resolutionselect分辨率 · 默认值:720p · 选项:480p、720p
  • variants: veed-fabric-1.0:flow
  • resolution: 480p | 720p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"veed-fabric-1.0","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","audio_url":"https://example.com/input.mp3","resolution":"720p"}}'
ltx-2.5-audio-to-video

LTX-2.5 Audio-to-Video Pro:根据你的音频(语音、音乐)生成具有电影感的 1080p 视频,并与音频同步:照片作为首帧,prompt 描述如何对其进行动画处理(没有照片时必须提供 prompt)。音频时长为 2–10 秒。它不是专用的唇形同步模型:与头像模型相比,镜头和姿势的移动更加自由。按输入音频的秒数计费。

变体(addressAs)
参数类型描述
ltx-2.5-audio-to-video:flowflowFlow · 可用
opts 参数
参数类型描述
image_urltext图像(URL)
audio_urlreq.text音频(URL)
aspect_ratioselect宽高比 · 默认值:auto · 选项:auto、16:9、9:16
guidance_scalenumber提示词遵循度(1–50)
  • variants: ltx-2.5-audio-to-video:flow
  • aspect_ratio: auto | 16:9 | 9:16

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"ltx-2.5-audio-to-video","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"audio_url":"https://example.com/input.mp3","aspect_ratio":"auto"}}'
pika

Pika 2.2 — 使用 Flow 将图像转换为视频。720p/1080p,5s。

变体(addressAs)
参数类型描述
pikaflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
seednumberSeed
negative_prompttext负面提示词
durationselect时长,秒 · 默认值:5 · 选项:5
  • 变体:pika
  • resolution: 720p | 1080p (affects price)
  • duration: 5

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"pika","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"image_url":"https://example.com/input.jpg","resolution":"720p","duration":"5"}}'
p-video-avatar

通过 Dune 提供的 PrunaAI P-Video Avatar。使用肖像和音频或语音脚本生成会说话的头像。

变体(addressAs)
参数类型描述
p-video-avatarduneDune · 可用
opts 参数
参数类型描述
image_urltext肖像(URL)
audio_urltext音频(URL)
voice_scripttext声音脚本
resolutionselect分辨率 · 默认值:720p · 选项:720p、1080p
  • 变体:p-video-avatar
  • resolution: 720p | 1080p

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"p-video-avatar","prompt":"A cat on a skateboard, slow motion, sunset.","opts":{"resolution":"720p"}}'
suno-music-video

Suno create-music-video — 为你之前的某条 Suno 音轨渲染视频。source_task_id 是 Suno 任务 ID,source_track 是音轨编号(默认为 0)。视频使用与音轨相同的通道:suno:white 音轨会获得 suno-music-video:white,Pixel 音轨会获得 Pixel。每次请求按固定价格计费;片段长度遵循音轨长度,不会改变价格。author/domain_name 字幕仅适用于 Pixel。

变体(addressAs)
参数类型描述
suno-music-videopixelPixel · 可用
suno-music-video:whitewhiteWhite · Available
opts 参数
参数类型描述
source_task_idreq.text你的 Suno 作业 ID(jobId)
source_tracknumber结果中的音轨编号(0、1、……)
source_audio_idtext源音轨 id(audioId)
authortext作者(在片段中署名)
domain_nametext域名(在片段中标注)
  • variants: suno-music-video, suno-music-video:white
  • 层级:每次请求

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-music-video","prompt":"A cat on a skateboard, slow motion, sunset."}'

音频(TTS)(46)

model输入单位变体opts 参数
qwen3-ttstext每 1000 个字符qwen3-tts, qwen3-tts:flowvoice, language, instruct, response_format
gpt-4o-mini-ttstext每 1000 个字符gpt-4o-mini-ttsvoice, instructions, response_format, speed
elevenlabs-v4-ttstext每 1000 个字符elevenlabs-v4-ttsengine, voice, voice_id, language_code, stability, similarity_boost, apply_text_normalization, output_format, seed
grok-ttstext每 1000 个字符grok-ttsvoice, language
elevenlabs-sound-effects-v2text每个文件elevenlabs-sound-effects-v2duration, prompt_influence, loop, output_format
mirelo-sfx-2text每个文件mirelo-sfx-2duration, negative_prompt, output_format, seed
mirelo-sfx-2-videotext, video每秒mirelo-sfx-2-videovideo_url, preserve_speech, negative_prompt, output_format, seed
ttstext每段旁白—engine, voice, language_code, minimax_voice, qwen3_voice, gemini_voice, stability, similarity_boost, speed, style, timestamps, previous_text, next_text, voice_id
elevenlabs-dialoguetext每 1000 个字符—dialogue, stability
gemini-ttstext每 1000 个字符—engine, gemini_voice, accent, style, pace, audio_profile, scene, sample_context, temperature
gemini-3.8-flash-ttstext每 1000 个字符gemini-3.8-flash-tts, gemini-3.8-flash-tts:flowengine, gemini_voice, style_instructions
gemini-3.8-flash-tts-litetext每 1000 个字符gemini-3.8-flash-tts-lite, gemini-3.8-flash-tts-lite:flowgemini_voice, style_instructions
chatterbox-ttstext每 1000 个字符chatterbox-tts, chatterbox-tts:flowvoice, exaggeration, temperature, seed
kokoro-ttstext每 1000 个字符kokoro-tts, kokoro-tts:flowvoice, speed
speech-to-textaudio每分钟speech-to-text:flow, speech-to-text:elevenlabsaudio_url, duration_minutes, language_code, diarize, tag_audio_events, keyterms
elevenlabs-audio-isolationaudio每分钟—audio_url, duration_minutes
nemotron-3-diarizationaudio每分钟nemotron-3-diarizationaudio_url, duration_minutes, language_code
qwen3-asr-1.7baudio每分钟qwen3-asr-1.7baudio_url, duration_minutes, language_code
qwen3-asr-0.6baudio每分钟qwen3-asr-0.6baudio_url, duration_minutes, language_code
voxtral-small-24b-2507audio每分钟voxtral-small-24b-2507audio_url, duration_minutes, language_code
mimo-v2-5-ttstext每 1000 个字符mimo-v2-5-ttsvoice, instruct, output_format
mimo-v2-5-tts-voiceclonetext每 1000 个字符mimo-v2-5-tts-voiceclonevoice, instruct
mimo-v2-5-tts-voicedesigntext每 1000 个字符mimo-v2-5-tts-voicedesignvoice
inworld-realtime-tts-1-5-minitext每 1000 个字符inworld-realtime-tts-1-5-minivoice, speaking_rate
inworld-realtime-tts-1-5-maxtext每 1000 个字符inworld-realtime-tts-1-5-maxvoice, speaking_rate, return_timestamps
whisper-large-v3-turboaudio每分钟whisper-large-v3-turboaudio_url, duration_minutes, language_code
nemotron-asr-multilingualaudio每分钟nemotron-asr-multilingualaudio_url, duration_minutes, language_code
whisper-large-v3audio每分钟whisper-large-v3audio_url, duration_minutes, language_code
gpt-4o-transcribeaudio每分钟gpt-4o-transcribeaudio_url, duration_minutes, language_code
gpt-4o-mini-transcribeaudio每分钟gpt-4o-mini-transcribeaudio_url, duration_minutes, language_code
gpt-4o-transcribe-diarizeaudio每分钟gpt-4o-transcribe-diarizeaudio_url, duration_minutes, language_code
universal-2audio每分钟universal-2audio_url, duration_minutes, language_code, speaker_labels, keyterms_prompt
forced-alignmenttext, audio每 1000 个字符—audio_url
chatterbox-turbotext每 1000 个字符chatterbox-turboprompt
voice-clone-ttstext每 1000 个字符—voice_id, engine, language, instruct, stability, similarity_boost, seed, previous_text, next_text, apply_text_normalization, previous_request_ids, next_request_ids, emotion, speed, pitch, volume, pronunciation_dict, voice_modify_pitch, voice_modify_timbre, voice_modify_intensity, response_format, sample_rate, bitrate, channel
elevenlabs-voice-changeraudio每 1000 个字符—voice_id, model_version, audio_url, similarity_boost, stability, remove_background_noise, response_format, sample_rate, seed
elevenlabs-dubbingaudio, video每秒elevenlabs-dubbingvideo_url, audio_url, target_language, source_language, cloning_strength
suno-stemaudio每个文件—audio_url, type, stem_name
suno-sfxtext每个文件—version, sound_loop, sound_tempo, sound_key
suno-lyricstext每个文件—prompt
suno-wavaudio每个文件suno-wav, suno-wav:whitesource_task_id, source_track, source_audio_id
suno-timestamped-lyricsaudio每个文件—source_task_id, source_track, source_audio_id
suno-style-boosttext每个文件—prompt
suno-personaaudio每个文件—source_task_id, source_track, source_audio_id, persona_name, persona_description
suno-midiaudio每个文件—source_task_id, source_track, source_audio_id
suno-speechtext每段旁白—mode, script, tone, backgroundMusic, vocalGender, variety
qwen3-tts

Qwen3-TTS (Alibaba) — 支持包括俄语在内的 10 种语言的语音合成:提供 9 种预设声音,并支持用普通文字控制语调。每次请求最多 4000 个字符。按每 1000 个字符计费。

变体(addressAs)
参数类型描述
qwen3-ttsduneDune · 可用
qwen3-tts:flowflowFlow · 可用
opts 参数
参数类型描述
voiceselect声音 · 默认值:Vivian · 选项:Vivian、Serena、Uncle_Fu、Dylan、Eric、Ryan、Aiden、Ono_Anna、Sohee
languageselect语言 · 默认值:auto · 选项:auto、ru、en、zh、ja、ko、de、fr、pt、es、it
instructtext音调指令(例如 "calm, whispering")
response_formatselect格式 · 默认值:mp3 · 选项:mp3、wav、opus、flac
  • variants: qwen3-tts, qwen3-tts:flow
  • voice: Vivian | Serena | Uncle_Fu | Dylan | Eric | Ryan | Aiden | Ono_Anna | Sohee
  • language: auto | ru | en | zh | ja | ko | de | fr | pt | es | it
  • response_format: mp3 | wav | opus | flac

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"voice":"Vivian","language":"auto","response_format":"mp3"}}'
gpt-4o-mini-tts

OpenAI gpt-4o-mini-tts — 支持包括俄语在内的多种语言的语音合成:13 种声音,并可使用自然语言控制表达方式(指令:语调、语速、情感)。每次请求最多 4096 个字符。兼容 OpenAI POST /v1/audio/speech。按每 1000 个字符计费。

变体(addressAs)
参数类型描述
gpt-4o-mini-ttswhiteWhite · Available
opts 参数
参数类型描述
voiceselect声音 · 默认值:alloy · 选项:alloy、ash、ballad、coral、echo、fable、nova、onyx、sage、shimmer、verse、marin、cedar
instructionstext传递说明(例如“平静、友好”)
response_formatselect格式 · 默认值:mp3 · 选项:mp3、opus、aac、flac、wav、pcm
speednumberSpeed (0.25–4)
  • variants: gpt-4o-mini-tts
  • voice: alloy | ash | ballad | coral | echo | fable | nova | onyx | sage | shimmer | verse | marin | cedar
  • response_format: mp3 | opus | aac | flac | wav | pcm

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-mini-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"voice":"alloy","response_format":"mp3"}}'
elevenlabs-v4-tts

ElevenLabs Eleven v4 — ElevenLabs 最新的语音模型(2026 年 9 月):富有表现力的表达方式、内联情绪标签([whispering]、[excited]),支持包括俄语在内的多种语言。Turbo 引擎价格减半且速度更快。每次请求最多 5000 个字符。按每 1000 个字符定价,并根据文本长度按比例计费。

变体(addressAs)
参数类型描述
elevenlabs-v4-ttsflowFlow · 可用
opts 参数
参数类型描述
engineselect引擎 · 默认值:v4 · 选项:v4、turbo
voiceselect声音 · 默认值:Rachel · 选项:Rachel、Aria、Roger、Sarah、Laura、Charlie、George、Callum、River、Liam、Charlotte、Alice、Matilda、Will、Jessica、Eric、Chris、Brian、Daniel、Lily、Bill
voice_idtext声音 ID(您自己的 ElevenLabs 声音,会覆盖列表)
language_codeselect语言 · 默认值:auto · 选项:auto、ru、en、de、fr、es、it、pt、pl、uk、tr、zh、ja、ko
stabilitynumberStability (0–1)
similarity_boostnumber相似度增强(0–1)
apply_text_normalizationselect规范化数字和日期 · 默认值:auto · 选项:auto、on、off
output_formatselect格式 · 默认值:mp3 · 选项:mp3、mp3_hq、opus
seednumberSeed(可选)
  • variants: elevenlabs-v4-tts
  • engine: v4 | turbo
  • voice: Rachel | Aria | Roger | Sarah | Laura | Charlie | George | Callum | River | Liam | Charlotte | Alice | Matilda | Will | Jessica | Eric | Chris | Brian | Daniel | Lily | Bill
  • language_code: auto | ru | en | de | fr | es | it | pt | pl | uk | tr | zh | ja | ko
  • apply_text_normalization: auto | on | off
  • output_format: mp3 | mp3_hq | opus

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"elevenlabs-v4-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"engine":"v4","voice":"Rachel","language_code":"auto","apply_text_normalization":"auto","output_format":"mp3"}}'
grok-tts

xAI Grok TTS — xAI 的语音合成:28 种声音,支持包括俄语在内的 20 种语言。每次请求最多 15000 个字符。按每 1000 个字符计费,并根据文本长度按比例收费。

变体(addressAs)
参数类型描述
grok-ttsflowFlow · 可用
opts 参数
参数类型描述
voiceselect声音 · 默认值:eve · 选项:eve、ara、rex、sal、leo、carina、zagan、helix、orion、luna、iris、altair、zenith、perseus、helios、lux、kepler、rigel、cosmo、celeste、ursa、sirius、lumen、castor、naksh、atlas、aurora、liora
languageselect语言 · 默认值:auto · 选项:auto、ru、en、de、fr、es-ES、es-MX、it、pt-BR、pt-PT、tr、zh、ja、ko、hi、bn、id、vi、ar-EG、ar-SA、ar-AE
  • 变体:grok-tts
  • voice: eve | ara | rex | sal | leo | carina | zagan | helix | orion | luna | iris | altair | zenith | perseus | helios | lux | kepler | rigel | cosmo | celeste | ursa | sirius | lumen | castor | naksh | atlas | aurora | liora
  • language: auto | ru | en | de | fr | es-ES | es-MX | it | pt-BR | pt-PT | tr | zh | ja | ko | hi | bn | id | vi | ar-EG | ar-SA | ar-AE

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"grok-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"voice":"eve","language":"auto"}}'
elevenlabs-sound-effects-v2

ElevenLabs Sound Effects V2:根据文本描述生成音效,包括脚步声、门声、爆炸声、自然声音、UI 音效和环境音垫。描述最多 450 个字符。时长为 0.5 到 22 秒,并可选择无缝循环。价格取决于时长。

变体(addressAs)
参数类型描述
elevenlabs-sound-effects-v2flowFlow · 可用
opts 参数
参数类型描述
durationselectDuration, sec · default: 5 · options: 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22
prompt_influencenumber提示词影响力(0–1)· 默认值:0.3
loopbool无缝循环 · 默认值:false
output_formatselect格式 · 默认值:mp3 · 选项:mp3、mp3_hq
  • variants: elevenlabs-sound-effects-v2
  • duration: 0.5 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22
  • output_format: mp3 | mp3_hq

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"elevenlabs-sound-effects-v2","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"duration":"5","prompt_influence":"0.3","output_format":"mp3"}}'
mirelo-sfx-2

Mirelo SFX 2.0 — 根据文本描述生成音效和环境音(最多 2000 个字符,英语效果最佳):脚步声、门、机器、自然声、城市噪音。时长为 1 至 60 秒;可以列出需要排除的声音。按声音秒数计费。

变体(addressAs)
参数类型描述
mirelo-sfx-2flowFlow · 可用
opts 参数
参数类型描述
durationselectDuration, sec · default: 5 · options: 1, 2, 3, 5, 8, 10, 15, 20, 30, 45, 60
negative_prompttext需要排除的声音
output_formatselect格式 · 默认值:mp3 · 选项:mp3、wav
seednumberSeed(可选)
  • variants: mirelo-sfx-2
  • duration: 1 | 2 | 3 | 5 | 8 | 10 | 15 | 20 | 30 | 45 | 60
  • output_format: mp3 | wav

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"mirelo-sfx-2","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"duration":"5","output_format":"mp3"}}'
mirelo-sfx-2-video

Mirelo SFX 2.0 for video — 为视频片段生成声音:模型会观看视频,并为其整个时长创建同步的音效和环境音(视频最长 60 秒)。可以保留源音轨中的语音。提示词为可选项,可用于提示需要听到的内容。结果为音频文件。按视频秒数计费。

变体(addressAs)
参数类型描述
mirelo-sfx-2-videoflowFlow · 可用
opts 参数
参数类型描述
video_urlreq.text视频(URL)
preserve_speechbool保留视频中的语音 · 默认值:true
negative_prompttext需要排除的声音
output_formatselect格式 · 默认值:mp3 · 选项:mp3、wav
seednumberSeed(可选)
  • variants: mirelo-sfx-2-video
  • output_format: mp3 | wav

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"mirelo-sfx-2-video","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"video_url":"https://example.com/input.mp4","output_format":"mp3"}}'
tts

使用 MiniMax、Qwen、Gemini 和 ElevenLabs 进行语音合成。每 1000 个字符的价格。支持俄语、声音选择和合成控制。

opts 参数
参数类型描述
engineselect引擎 · 默认值:minimax-turbo · 选项:turbo、multilingual、minimax-turbo、minimax-hd、qwen3、gemini-flash
voiceselect声音 · 默认值:EkK5I93UQWFDigLMpZcX · 选项:EkK5I93UQWFDigLMpZcX、Xb7hH8MSUJpSbSDYk0k2、pNInz6obpgDQGcFmaJgB
language_codeselect语言 · 默认值:auto · 选项:auto、ru、en
minimax_voiceselectMiniMax voice · default: Wise_Woman · options: Wise_Woman, Friendly_Person, Inspirational_girl, Deep_Voice_Man, Calm_Woman, Casual_Guy, Lively_Girl, Patient_Man, Young_Knight, Determined_Man, Lovely_Girl, Decent_Boy, Imposing_Manner, Elegant_Man, Abbess, Sweet_Girl_2, Exuberant_Girl
qwen3_voiceselectQwen 语音 · 默认值:Vivian · 选项:Vivian、Serena、Uncle_Fu、Dylan、Eric、Ryan、Aiden、Ono_Anna、Sohee
gemini_voiceselectGemini voice · 默认值:Kore · 选项:Achernar、Achird、Algenib、Algieba、Alnilam、Aoede、Autonoe、Callirrhoe、Charon、Despina、Enceladus、Erinome、Fenrir、Gacrux、Iapetus、Kore、Laomedeia、Leda、Orus、Pulcherrima、Puck、Rasalgethi、Sadachbia、Sadaltager、Schedar、Sulafat、Umbriel、Vindemiatrix、Zephyr、Zubenelgenubi
stabilitynumberStability (0–1)
similarity_boostnumber相似度增强(0–1)
speednumberSpeed (0.7–1.2)
stylenumberStyle (0–1)
timestampsselect时间戳 · 默认值:false · 选项:false, true
previous_texttext之前的文本(最多 5000)
next_texttext下一段文本(最多 5000)
voice_idtext声音 ID(自定义声音)
  • engine: turbo | multilingual | minimax-turbo | minimax-hd | qwen3 | gemini-flash (affects price)
  • voice: EkK5I93UQWFDigLMpZcX | Xb7hH8MSUJpSbSDYk0k2 | pNInz6obpgDQGcFmaJgB
  • language_code: auto | ru | en
  • minimax_voice: Wise_Woman | Friendly_Person | Inspirational_girl | Deep_Voice_Man | Calm_Woman | Casual_Guy | Lively_Girl | Patient_Man | Young_Knight | Determined_Man | Lovely_Girl | Decent_Boy | Imposing_Manner | Elegant_Man | Abbess | Sweet_Girl_2 | Exuberant_Girl
  • qwen3_voice: Vivian | Serena | Uncle_Fu | Dylan | Eric | Ryan | Aiden | Ono_Anna | Sohee
  • gemini_voice: Achernar | Achird | Algenib | Algieba | Alnilam | Aoede | Autonoe | Callirrhoe | Charon | Despina | Enceladus | Erinome | Fenrir | Gacrux | Iapetus | Kore | Laomedeia | Leda | Orus | Pulcherrima | Puck | Rasalgethi | Sadachbia | Sadaltager | Schedar | Sulafat | Umbriel | Vindemiatrix | Zephyr | Zubenelgenubi
  • timestamps: false | true

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"engine":"minimax-turbo","language_code":"auto","minimax_voice":"Wise_Woman"}}'
elevenlabs-dialogue

ElevenLabs V3 多人声对话合成。提供一个对话数组,为每行指定文本和声音。按比例计费,价格为每 1000 个字符 $0.10。

opts 参数
参数类型描述
dialoguereq.array对话行({text, voice} 对象)
stabilitynumberStability (0–1)
  • 层级:每 1000 个字符

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"elevenlabs-dialogue","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
gemini-tts

Google Gemini TTS (Pixel) — 对话式语音合成(speakers + dialogue_turns)。

opts 参数
参数类型描述
engineselect引擎 · 默认值:flash · 选项:flash、pro
gemini_voiceselectGemini voice · 默认值:Kore · 选项:Achernar、Achird、Algenib、Algieba、Alnilam、Aoede、Autonoe、Callirrhoe、Charon、Despina、Enceladus、Erinome、Fenrir、Gacrux、Iapetus、Kore、Laomedeia、Leda、Orus、Pulcherrima、Puck、Rasalgethi、Sadachbia、Sadaltager、Schedar、Sulafat、Umbriel、Vindemiatrix、Zephyr、Zubenelgenubi
accentselect口音 · 默认值:Neutral · 可选值:Neutral、American (Gen)、American (Valley)、American (South)、British (RP)、British (Brixton)、Transatlantic、Australian
styleselect风格 · 默认值:Vocal Smile · 可选值:Vocal Smile、Newscaster、Whisper、Empathetic、Promo/Hype、Deadpan
paceselect节奏 · 默认值:Natural · 选项:Natural、Rapid Fire、The Drift、Staccato
audio_profiletext音频配置文件(语音描述)
scenetext场景描述
sample_contexttext示例上下文 / 语气
temperaturenumberTemperature (0–2)
  • engine: flash | pro (affects price)
  • gemini_voice: Achernar | Achird | Algenib | Algieba | Alnilam | Aoede | Autonoe | Callirrhoe | Charon | Despina | Enceladus | Erinome | Fenrir | Gacrux | Iapetus | Kore | Laomedeia | Leda | Orus | Pulcherrima | Puck | Rasalgethi | Sadachbia | Sadaltager | Schedar | Sulafat | Umbriel | Vindemiatrix | Zephyr | Zubenelgenubi
  • accent: Neutral | American (Gen) | American (Valley) | American (South) | British (RP) | British (Brixton) | Transatlantic | Australian
  • style: Vocal Smile | Newscaster | Whisper | Empathetic | Promo/Hype | Deadpan
  • pace: Natural | Rapid Fire | The Drift | Staccato

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"engine":"flash","gemini_voice":"Kore","accent":"Neutral","style":"Vocal Smile","pace":"Natural"}}'
gemini-3.8-flash-tts

Google Gemini 3.8 Flash TTS — 新一代语音合成:30 种声音,支持多语言并自动检测语言。Flash Lite 引擎价格更低。按 1000 个字符计费。

变体(addressAs)
参数类型描述
gemini-3.8-flash-ttspixelPixel · 可用
gemini-3.8-flash-tts:flowflowFlow · 可用
opts 参数
参数类型描述
engineselect引擎 · 默认值:flash · 选项:flash、lite
gemini_voiceselectGemini voice · 默认值:Kore · 选项:Achernar、Achird、Algenib、Algieba、Alnilam、Aoede、Autonoe、Callirrhoe、Charon、Despina、Enceladus、Erinome、Fenrir、Gacrux、Iapetus、Kore、Laomedeia、Leda、Orus、Pulcherrima、Puck、Rasalgethi、Sadachbia、Sadaltager、Schedar、Sulafat、Umbriel、Vindemiatrix、Zephyr、Zubenelgenubi
style_instructionstext传递风格(说明)
  • variants: gemini-3.8-flash-tts, gemini-3.8-flash-tts:flow
  • engine: flash | lite (affects price)
  • gemini_voice: Achernar | Achird | Algenib | Algieba | Alnilam | Aoede | Autonoe | Callirrhoe | Charon | Despina | Enceladus | Erinome | Fenrir | Gacrux | Iapetus | Kore | Laomedeia | Leda | Orus | Pulcherrima | Puck | Rasalgethi | Sadachbia | Sadaltager | Schedar | Sulafat | Umbriel | Vindemiatrix | Zephyr | Zubenelgenubi

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-3.8-flash-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"engine":"flash","gemini_voice":"Kore"}}'
gemini-3.8-flash-tts-lite

Google Gemini 3.8 Flash Lite TTS — 更轻量、更低价的新一代语音合成:30 种声音,支持多语言并自动检测语言。按 1000 个字符计费。

变体(addressAs)
参数类型描述
gemini-3.8-flash-tts-litepixelPixel · 可用
gemini-3.8-flash-tts-lite:flowflowFlow · 可用
opts 参数
参数类型描述
gemini_voiceselectGemini voice · 默认值:Kore · 选项:Achernar、Achird、Algenib、Algieba、Alnilam、Aoede、Autonoe、Callirrhoe、Charon、Despina、Enceladus、Erinome、Fenrir、Gacrux、Iapetus、Kore、Laomedeia、Leda、Orus、Pulcherrima、Puck、Rasalgethi、Sadachbia、Sadaltager、Schedar、Sulafat、Umbriel、Vindemiatrix、Zephyr、Zubenelgenubi
style_instructionstext传递风格(说明)
  • variants: gemini-3.8-flash-tts-lite, gemini-3.8-flash-tts-lite:flow
  • gemini_voice: Achernar | Achird | Algenib | Algieba | Alnilam | Aoede | Autonoe | Callirrhoe | Charon | Despina | Enceladus | Erinome | Fenrir | Gacrux | Iapetus | Kore | Laomedeia | Leda | Orus | Pulcherrima | Puck | Rasalgethi | Sadachbia | Sadaltager | Schedar | Sulafat | Umbriel | Vindemiatrix | Zephyr | Zubenelgenubi (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-3.8-flash-tts-lite","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"gemini_voice":"Kore"}}'
chatterbox-tts

Resemble AI Chatterbox,多语言 TTS,支持 23 种语言。按每 1000 个字符计费。限制:每个请求最多 300 个字符。

变体(addressAs)
参数类型描述
chatterbox-ttsduneDune · 可用
chatterbox-tts:flowflowFlow · 可用
opts 参数
参数类型描述
voiceselect语言 · 默认值:english · 选项:english、arabic、danish、german、greek、spanish、finnish、french、hebrew、hindi、italian、japanese、korean、malay、dutch、norwegian、polish、portuguese、russian、swedish、swahili、turkish、chinese
exaggerationnumberExpressiveness (0.25–1)
temperaturenumberTemperature (0.05–2)
seednumberSeed
  • 变体:chatterbox-tts、chatterbox-tts:flow
  • voice: english | arabic | danish | german | greek | spanish | finnish | french | hebrew | hindi | italian | japanese | korean | malay | dutch | norwegian | polish | portuguese | russian | swedish | swahili | turkish | chinese (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"chatterbox-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"voice":"english"}}'
kokoro-tts

Kokoro TTS:轻量、快速的美式英语 TTS。仅支持英语。

变体(addressAs)
参数类型描述
kokoro-ttsduneDune · 可用
kokoro-tts:flowflowFlow · 可用
opts 参数
参数类型描述
voiceselectVoice · default: af_heart · options: af_heart, af_alloy, af_aoede, af_bella, af_jessica, af_kore, af_nicole, af_nova, af_river, af_sarah, af_sky, am_adam, am_echo, am_eric, am_fenrir, am_liam, am_michael, am_onyx, am_puck, am_santa
speednumberSpeed (0.25–4)
  • 变体:kokoro-tts、kokoro-tts:flow
  • voice: af_heart | af_alloy | af_aoede | af_bella | af_jessica | af_kore | af_nicole | af_nova | af_river | af_sarah | af_sky | am_adam | am_echo | am_eric | am_fenrir | am_liam | am_michael | am_onyx | am_puck | am_santa (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kokoro-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"voice":"af_heart"}}'
speech-to-text

通过 Flow 或 ElevenLabs Scribe v2 进行语音识别。按输入音频的分钟数计价,并根据测量时长计费。

变体(addressAs)
参数类型描述
speech-to-text:flowflowFlow · 可用
speech-to-text:elevenlabselevenlabsElevenLabs · Available
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
diarizebool说话人分离 · 默认值:true
tag_audio_eventsbool音频事件标签 · 默认值:true
keytermstext关键术语(以逗号分隔,Flow 费率 +30%)
  • variants: speech-to-text:flow, speech-to-text:elevenlabs

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"speech-to-text","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
elevenlabs-audio-isolation

ElevenLabs Voice Isolator — 从语音中去除背景噪音和音乐。按输入音频的分钟数计价;最短时长为 4.6 秒。

opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"elevenlabs-audio-isolation","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
nemotron-3-diarization

通过 Dune 提供的 NVIDIA Nemotron diarization。支持带有说话人片段和时间戳的语音识别。

变体(addressAs)
参数类型描述
nemotron-3-diarizationduneDune · 可用
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • variants: nemotron-3-diarization

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"nemotron-3-diarization","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
qwen3-asr-1.7b

通过 Dune 提供的 Qwen3-ASR 1.7B。使用 multipart 音频输入进行语音识别和翻译。

变体(addressAs)
参数类型描述
qwen3-asr-1.7bduneDune · 可用
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • variants: qwen3-asr-1.7b

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3-asr-1.7b","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
qwen3-asr-0.6b

通过 Dune 提供的 Qwen3-ASR 0.6B。成本最低的语音识别:支持包括俄语在内的 30 种语言,提供分段时间戳;支持 multipart 音频输入。

变体(addressAs)
参数类型描述
qwen3-asr-0.6bduneDune · 可用
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • variants: qwen3-asr-0.6b

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3-asr-0.6b","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
voxtral-small-24b-2507

通过 Dune 提供的 Mistral Voxtral Small 24B 2507。通过音频接口进行语音理解和转录。

变体(addressAs)
参数类型描述
voxtral-small-24b-2507duneDune · 可用
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • variants: voxtral-small-24b-2507

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"voxtral-small-24b-2507","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
mimo-v2-5-tts

通过 Dune 提供的 Xiaomi MiMo V2.5 TTS。支持选择声音和设置风格说明,输出格式为 WAV、MP3 或 Opus。

变体(addressAs)
参数类型描述
mimo-v2-5-ttsduneDune · 可用
opts 参数
参数类型描述
voicetext声音
instructtext指令
output_formatselect格式 · 默认值:wav · 选项:wav、mp3、opus
  • variants: mimo-v2-5-tts
  • output_format: wav | mp3 | opus

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"mimo-v2-5-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"output_format":"wav"}}'
mimo-v2-5-tts-voiceclone

通过 Dune 提供的 Xiaomi MiMo V2.5 声音克隆。使用您的声音样本 ID 进行合成。

变体(addressAs)
参数类型描述
mimo-v2-5-tts-voicecloneduneDune · 可用
opts 参数
参数类型描述
voicetext声音 ID
instructtext指令
  • variants: mimo-v2-5-tts-voiceclone

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"mimo-v2-5-tts-voiceclone","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
mimo-v2-5-tts-voicedesign

通过 Dune 提供的 Xiaomi MiMo V2.5 声音设计。用文本描述声音的音色和表达方式。

变体(addressAs)
参数类型描述
mimo-v2-5-tts-voicedesignduneDune · 可用
opts 参数
参数类型描述
voicetext声音描述
  • variants: mimo-v2-5-tts-voicedesign

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"mimo-v2-5-tts-voicedesign","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
inworld-realtime-tts-1-5-mini

通过 Dune 提供的 Inworld Realtime TTS 1.5 Mini。快速语音合成,支持俄语。

变体(addressAs)
参数类型描述
inworld-realtime-tts-1-5-miniduneDune · 可用
opts 参数
参数类型描述
voicetext声音
speaking_ratenumber语速(0.5–1.5)
  • variants: inworld-realtime-tts-1-5-mini

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"inworld-realtime-tts-1-5-mini","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
inworld-realtime-tts-1-5-max

通过 Dune 提供的 Inworld Realtime TTS 1.5 Max。最高质量,可选词级时间戳。

变体(addressAs)
参数类型描述
inworld-realtime-tts-1-5-maxduneDune · 可用
opts 参数
参数类型描述
voicetext声音
speaking_ratenumber语速(0.5–1.5)
return_timestampsbool单词时间戳
  • variants: inworld-realtime-tts-1-5-max

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"inworld-realtime-tts-1-5-max","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
whisper-large-v3-turbo

通过 Dune 提供的 OpenAI Whisper large-v3-turbo。按输入音频分钟数计费。

变体(addressAs)
参数类型描述
whisper-large-v3-turboduneDune · 可用
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • variants: whisper-large-v3-turbo

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"whisper-large-v3-turbo","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
nemotron-asr-multilingual

通过 Dune 提供的 NVIDIA Nemotron 3.5 ASR Streaming Multilingual 0.6b。按输入音频分钟数计费。

变体(addressAs)
参数类型描述
nemotron-asr-multilingualduneDune · 可用
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • 变体:nemotron-asr-multilingual

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"nemotron-asr-multilingual","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
whisper-large-v3

通过 Dune 提供的 OpenAI Whisper large-v3。按输入音频分钟数计费。

变体(addressAs)
参数类型描述
whisper-large-v3duneDune · 可用
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • variants: whisper-large-v3

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"whisper-large-v3","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
gpt-4o-transcribe

OpenAI gpt-4o-transcribe — 语音识别。卡片上的价格为每分钟费用上限;费用根据实际响应 token 数计算,因此通常更低,且绝不会超过上限。最低收费为一戈比(在 getvibeapi.com 上为一微美元)。支持识别俄语及其他语言;可指定语言,也可让模型自动检测。仅返回文本(json/text),不含时间戳。

变体(addressAs)
参数类型描述
gpt-4o-transcribewhiteWhite · Available
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • variants: gpt-4o-transcribe

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-transcribe","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
gpt-4o-mini-transcribe

OpenAI gpt-4o-mini-transcribe — 快速、低成本的语音识别。卡片上的价格为每分钟费用上限;费用根据实际响应 token 数计算,因此通常更低,且绝不会超过上限。最低收费为一戈比(在 getvibeapi.com 上为一微美元)。支持识别俄语及其他语言;可指定语言,也可让模型自动检测。仅返回文本(json/text),不含时间戳。

变体(addressAs)
参数类型描述
gpt-4o-mini-transcribewhiteWhite · Available
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • variants: gpt-4o-mini-transcribe

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-mini-transcribe","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
gpt-4o-transcribe-diarize

OpenAI gpt-4o-transcribe-diarize — 提供说话人标签(A、B、…)和语句时间戳的语音识别模型。卡片上的价格是每分钟费用上限;实际费用根据响应 token 数计算,因此通常低于上限,且绝不会超过上限。最低收费为 one kopeck(在 getvibeapi.com 上为 one micro-dollar)。支持识别俄语;请求时指定 response_format=diarized_json 或 verbose_json 即可获取说话人信息。

变体(addressAs)
参数类型描述
gpt-4o-transcribe-diarizeamberAmber · Available
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言
  • variants: gpt-4o-transcribe-diarize

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4o-transcribe-diarize","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
universal-2

Universal-2 — 俄语语音识别。language_code=ru 或自动检测;提供商契约选项(speaker_labels、keyterms_prompt 等)。按输入音频分钟数计费(最终扣款前估算)。

变体(addressAs)
参数类型描述
universal-2scribeScribe · 可用
opts 参数
参数类型描述
audio_urlreq.text音频(URL)
duration_minutesnumber价格预览时长,分钟(不用于计费)· 默认值:1
language_codetext语言(仅支持 ru;留空 = 自动检测)
speaker_labelsbool说话人分离 · 默认值:false
keyterms_prompttext关键术语(以逗号分隔)
  • variants: universal-2

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"universal-2","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","duration_minutes":"1"}}'
forced-alignment

强制对齐,获取精确的单词时间。发送一个已准备好的音频文件及其文字稿,即可获取每个单词的开始和结束时间:用于自动剪辑、修正时间点和制作字幕。适用于任何配音,不仅限于我们的配音。任务完成后会附带一个 SRT 文件。按所提交文本的每 1000 个字符计费。

opts 参数
参数类型描述
audio_urlreq.text音频(URL)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"forced-alignment","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3"}}'
chatterbox-turbo

通过 Dune 提供的 ResembleAI Chatterbox Turbo。按输入字符数计费。

变体(addressAs)
参数类型描述
chatterbox-turboduneDune · 可用
opts 参数
参数类型描述
promptreq.text用于语音合成的文本
  • 变体:chatterbox-turbo

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"chatterbox-turbo","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
voice-clone-tts

声音克隆:使用您自己的声音进行语音合成。上传一段您的声音短录音(3–60 秒,wav/mp3),通过 POST /api/media/voices 或在工作室中上传,获取 voice_id,然后使用它合成任意文本,包括俄语。创建标准声音免费。Premium 声音可提供最接近本人声音的效果:10–60 秒的录音,创建时一次性支付 $2.00,可使用 MiniMax Speech 02、2.6 和 2.8 引擎(HD 和 Turbo)进行合成。合成按每 1000 个字符计费。每次请求最多 4000 个字符。

opts 参数
参数类型描述
voice_idreq.text您的声音(voice_id)
engineselectEngine · default: qwen3 · options: qwen3, chatterbox, minimax-speech-02-hd, minimax-speech-02-turbo, minimax-speech-2.6-hd, minimax-speech-2.6-turbo, minimax-speech-2.8-hd, minimax-speech-2.8-turbo, elevenlabs-v4, elevenlabs-v4-turbo
languageselect语言 · 默认值:auto · 选项:auto、ru、en、de、fr、es、it、pt、zh、ja、ko
instructtext音调指令(例如 "calm")
stabilitynumberStability (0–1) · default: 0.5
similarity_boostnumberSimilarity to the sample (0–1) · default: 0.75
seednumberSeed (repeatable result)
previous_texttextText before (for smooth joins)
next_texttextText after (for smooth joins)
apply_text_normalizationselectSpell out numbers and dates · default: auto · options: auto, on, off
previous_request_idsarrayPrevious request ids (resultMeta.elevenlabsRequestId, up to 3)
next_request_idsarrayNext request ids (resultMeta.elevenlabsRequestId, up to 3)
emotionselect情绪 · 选项:、neutral、happy、sad、angry、fearful、disgusted、surprised
speednumberSpeech speed (0.5–2; ElevenLabs 0.7–1.2) · default: 1
pitchnumber音调(−12…12)· 默认值:0
volumenumber音量(0.01–10)· 默认值:1
pronunciation_dictarray发音词典(单词/读法;…)
voice_modify_pitchnumber音调偏移(−100…100)· 默认值:0
voice_modify_timbrenumber音色(−100…100)· 默认值:0
voice_modify_intensitynumber强度(−100…100) · 默认值:0
response_formatselectFormat · default: mp3 · options: mp3, wav, flac, opus, pcm
sample_rateselectSample rate · options: , 8000, 16000, 22050, 24000, 32000, 44100, 48000
bitrateselectMP3 bitrate · options: , 32000, 64000, 128000, 256000
channelselect通道 · 选项:、1、2
  • engine: qwen3 | chatterbox | minimax-speech-02-hd | minimax-speech-02-turbo | minimax-speech-2.6-hd | minimax-speech-2.6-turbo | minimax-speech-2.8-hd | minimax-speech-2.8-turbo | elevenlabs-v4 | elevenlabs-v4-turbo
  • language: auto | ru | en | de | fr | es | it | pt | zh | ja | ko
  • apply_text_normalization: auto | on | off
  • emotion: | neutral | happy | sad | angry | fearful | disgusted | surprised
  • response_format: mp3 | wav | flac | opus | pcm
  • sample_rate: | 8000 | 16000 | 22050 | 24000 | 32000 | 44100 | 48000
  • bitrate: | 32000 | 64000 | 128000 | 256000
  • channel: | 1 | 2

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"voice-clone-tts","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"engine":"qwen3","language":"auto","response_format":"mp3"}}'
elevenlabs-voice-changer

变声器:源录音使用您克隆的 ElevenLabs 声音进行朗读,同时保留原始录音的语调、停顿、耳语和笑声。需要使用 engine: elevenlabs 创建的声音(POST /api/media/voices)。源文件:最长 5 分钟的 wav/mp3。按每 1000 个字符计费:1 分钟源音频 = 1000 个字符。

opts 参数
参数类型描述
voice_idreq.text您的声音(voice_id)
model_versionselect模型版本 · 默认值:multilingual · 选项:multilingual、english
audio_urlreq.text源音频(URL)
similarity_boostnumber与样本的相似度(0–1)
stabilitynumberStability (0–1)
remove_background_noisebool移除背景噪音 · 默认值:false
response_formatselect格式 · 默认值:mp3 · 选项:mp3、wav、opus、pcm
sample_rateselectSample rate · options: , 8000, 16000, 22050, 24000, 32000, 44100, 48000
seednumberSeed(可选)
  • model_version: multilingual | english
  • response_format: mp3 | wav | opus | pcm
  • sample_rate: | 8000 | 16000 | 22050 | 24000 | 32000 | 44100 | 48000

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"elevenlabs-voice-changer","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"model_version":"multilingual","audio_url":"https://example.com/input.mp3","response_format":"mp3","sample_rate":""}}'
elevenlabs-dubbing

ElevenLabs Dubbing — 将音频或视频文件中的语音翻译成另一种语言,同时保留每位说话者的声音、情感和时间安排;支持 90+ 种语言。一个源文件:video_url(结果为配音视频)或 audio_url(结果为音频),最长 30 分钟。模型会区分说话者;源语言会自动检测,也可以通过 source_language 设置。按结果时长的秒数计费;时长取自源文件。

变体(addressAs)
参数类型描述
elevenlabs-dubbingcanvasCanvas · 可用
opts 参数
参数类型描述
video_urltext源视频(URL)
audio_urltext源音频(URL)
target_languagereq.select目标语言 · 默认值:en · 选项:af、ak、sq、am、ar、ar-EG、hy、as、az、eu、be、bs、bg、my、yue、ca、ceb、zh、zh-TW、hr、cs、da、dgo、nl、en、en-AU、en-CA、en-GB、en-US、et、fil、fi、fr、fr-CA、fr-FR、gl、ka、de、el、gu、ha、he、hi、hu、is、id、it、ja、jv、kn、kk、ki、rw、rn、ko、ky、lv、lt、lg、mk、ms、ml、cmn、mr、mn、ne、no、fa、pl、pt、pt-BR、pt-PT、pa、ro、ru、nso、st、sd、sk、sl、es、es-AR、es-CL、es-ES、es-MX、su、sw、ss、sv、tg、ta、te、th、bo、ts、tn、tr、uk、ur、ug、uz、ve、vi、war、cy、wo、yo、zu
source_languageselect源语言 · 默认值:auto · 选项:auto、af、ak、sq、am、ar、ar-EG、hy、as、az、eu、be、bs、bg、my、yue、ca、ceb、zh、zh-TW、hr、cs、da、dgo、nl、en、en-AU、en-CA、en-GB、en-US、et、fil、fi、fr、fr-CA、fr-FR、gl、ka、de、el、gu、ha、he、hi、hu、is、id、it、ja、jv、kn、kk、ki、rw、rn、ko、ky、lv、lt、lg、mk、ms、ml、cmn、mr、mn、ne、no、fa、pl、pt、pt-BR、pt-PT、pa、ro、ru、nso、st、sd、sk、sl、es、es-AR、es-CL、es-ES、es-MX、su、sw、ss、sv、tg、ta、te、th、bo、ts、tn、tr、uk、ur、ug、uz、ve、vi、war、cy、wo、yo、zu
cloning_strengthnumber与原始声音的相似度(0–10) · 默认值:7
  • 变体:elevenlabs-dubbing
  • target_language: af | ak | sq | am | ar | ar-EG | hy | as | az | eu | be | bs | bg | my | yue | ca | ceb | zh | zh-TW | hr | cs | da | dgo | nl | en | en-AU | en-CA | en-GB | en-US | et | fil | fi | fr | fr-CA | fr-FR | gl | ka | de | el | gu | ha | he | hi | hu | is | id | it | ja | jv | kn | kk | ki | rw | rn | ko | ky | lv | lt | lg | mk | ms | ml | cmn | mr | mn | ne | no | fa | pl | pt | pt-BR | pt-PT | pa | ro | ru | nso | st | sd | sk | sl | es | es-AR | es-CL | es-ES | es-MX | su | sw | ss | sv | tg | ta | te | th | bo | ts | tn | tr | uk | ur | ug | uz | ve | vi | war | cy | wo | yo | zu
  • source_language: auto | af | ak | sq | am | ar | ar-EG | hy | as | az | eu | be | bs | bg | my | yue | ca | ceb | zh | zh-TW | hr | cs | da | dgo | nl | en | en-AU | en-CA | en-GB | en-US | et | fil | fi | fr | fr-CA | fr-FR | gl | ka | de | el | gu | ha | he | hi | hu | is | id | it | ja | jv | kn | kk | ki | rw | rn | ko | ky | lv | lt | lg | mk | ms | ml | cmn | mr | mn | ne | no | fa | pl | pt | pt-BR | pt-PT | pa | ro | ru | nso | st | sd | sk | sl | es | es-AR | es-CL | es-ES | es-MX | su | sw | ss | sv | tg | ta | te | th | bo | ts | tn | tr | uk | ur | ug | uz | ve | vi | war | cy | wo | yo | zu

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"elevenlabs-dubbing","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"target_language":"en","source_language":"auto","cloning_strength":"7","video_url":"https://example.com/input.mp4"}}'
suno-stem

Suno:将音频拆分为人声和器乐音轨。

opts 参数
参数类型描述
audio_urlreq.text音频(URL)
typeselect分离模式 · 默认值:separate_vocal · 可选值:separate_vocal、split_stem、split_stem_advanced
stem_nametext要提取的音轨(高级)
  • type: separate_vocal | split_stem | split_stem_advanced (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-stem","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"audio_url":"https://example.com/input.mp3","type":"separate_vocal"}}'
suno-sfx

Suno:根据描述生成音效。

opts 参数
参数类型描述
versionselect版本 · 默认:v6 · 选项:v6、v6-wild、v6-mini
sound_loopbool循环播放声音 · 默认值:false
sound_temponumber节奏(BPM,1–300)
sound_keyselect调性 · 默认值:Any · 选项:Any、Cm、C#m、Dm、D#m、Em、Fm、F#m、Gm、G#m、Am、A#m、Bm、C、C#、D、D#、E、F、F#、G、G#、A、A#、B
  • version: v6 | v6-wild | v6-mini (affects price)
  • sound_key: Any | Cm | C#m | Dm | D#m | Em | Fm | F#m | Gm | G#m | Am | A#m | Bm | C | C# | D | D# | E | F | F# | G | G# | A | A# | B

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-sfx","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"version":"v6","sound_key":"Any"}}'
suno-lyrics

Suno:根据描述生成歌曲歌词。

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-lyrics","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
suno-wav

Suno — 生成你某条 Suno 音轨的 WAV 版本,或生成某个已分离音轨的 WAV 版本(suno-stem:人声、伴奏或单件乐器)。source_task_id 是 Suno 或 suno-stem 任务 ID,source_track 是其返回结果中的音轨或分离音轨编号(默认为 0)。suno:white 音轨会以 suno-wav:white 运行(从源任务中选择;分离音轨仅适用于 Pixel)。每次请求按固定价格计费。

变体(addressAs)
参数类型描述
suno-wavpixelPixel · 可用
suno-wav:whitewhiteWhite · Available
opts 参数
参数类型描述
source_task_idreq.text你的 Suno 作业 ID(jobId)
source_tracknumber结果中的音轨编号(0、1、……)
source_audio_idtext源音轨 id(audioId)
  • variants: suno-wav, suno-wav:white
  • 层级:每次请求

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-wav","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
suno-timestamped-lyrics

Suno TimeStamped Lyrics:为之前的 Suno 音轨提供逐词时间戳。同步返回结果。每次请求按固定价格计费。

opts 参数
参数类型描述
source_task_idreq.text你的 Suno 作业 ID(jobId)
source_tracknumber结果中的音轨编号(0、1、……)
source_audio_idtext源音轨 id(audioId)
  • 层级:每次请求

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-timestamped-lyrics","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
suno-style-boost

Suno Boost Music Style:将简短的风格描述扩展为更完整的提示词。同步返回结果。每次请求按固定价格计费。

  • 层级:Style Boost

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-style-boost","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
suno-persona

Suno Generate Persona:根据之前生成的 Suno 内容创建声音/风格角色。Pixel 将此路由发布为零积分,因此价格由所有者决定(2026-08-31)。

opts 参数
参数类型描述
source_task_idreq.text你的 Suno 作业 ID(jobId)
source_tracknumber结果中的音轨编号(0、1、……)
source_audio_idtext源音轨 id(audioId)
persona_namereq.textPersona 名称
persona_descriptionreq.textPersona 描述
  • 层级:每次请求

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-persona","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
suno-midi

Suno Generate Midi From Audio:根据之前生成的 Suno 内容生成 MIDI。Pixel 将此路由发布为零积分,因此价格由所有者决定(2026-08-31)。

opts 参数
参数类型描述
source_task_idreq.text你的 Suno 作业 ID(jobId)
source_tracknumber结果中的音轨编号(0、1、……)
source_audio_idtext源音轨 id(audioId)
  • 层级:每次请求

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-midi","prompt":"Hello! This is a test voiceover through the GetVibe API."}'
suno-speech

Suno Speech — 可控制语气、节奏、情感并可选配背景音乐的语音音频;并非固定音色的 TTS。简单模式根据一句场景描述编写并演绎脚本。高级模式演绎你提供的脚本(最多 5000 个字符,支持换行),并支持语气指导(最多 1000 个字符)、背景音乐、女声或男声以及 0–4 的变化度。每次请求返回两个音频片段。每次请求按固定价格计费。

opts 参数
参数类型描述
modeselect模式 · 默认值:simple · 选项:simple, advanced
scripttext脚本(最多 5000 个字符,支持换行)
tonetext语气与表达方式(最多 1000 个字符)
backgroundMusicbool背景音乐 · 默认值:false
vocalGenderselect声音 · 选项:, f, m
varietyselectVariety (0–4) · default: 0 · options: 0, 1, 2, 3, 4
  • 层级:每次请求
  • mode: simple | advanced
  • vocalGender: | f | m
  • variety: 0 | 1 | 2 | 3 | 4

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-speech","prompt":"Hello! This is a test voiceover through the GetVibe API.","opts":{"mode":"simple"}}'

Music (11)

model输入单位变体opts 参数
lyria-3-5text每个音轨lyria-3-5:white, lyria-3-5:flowlyrics, title, bpm, seed
elevenlabs-musictext每个音轨elevenlabs-musicduration, force_instrumental, output_format, seed
acestep-v15-xl-sfttext每秒acestep-v15-xl-sftlyrics, duration, instrumental, language_code
sunotext每个音轨suno, suno:white, suno:amber (unavailable)version, instrumental, style, title, negativeTags, vocalGender, styleWeight, weirdnessConstraint, audioWeight, personaId, personaModel, duration
suno-mashuptext, audio每个音轨—upload_urls, version, custom_mode, instrumental, style, title, vocalGender, styleWeight, weirdnessConstraint, audioWeight
suno-add-instrumentaltext, audio每个音轨—upload_url, version, title, tags, negativeTags, vocalGender, styleWeight, weirdnessConstraint, audioWeight
suno-add-vocalstext, audio每个音轨—upload_url, version, title, style, negativeTags, vocalGender, styleWeight, weirdnessConstraint, audioWeight
suno-cover-audiotext, audio每个音轨—upload_url, source_task_id, source_track, version, custom_mode, instrumental, style, title, negativeTags, vocalGender, styleWeight, weirdnessConstraint, audioWeight, personaId
suno-extend-audiotext, audio每个音轨—upload_url, version, continue_at, instrumental, style, title, negativeTags, vocalGender, styleWeight, weirdnessConstraint, audioWeight, personaId
suno-replace-sectiontext每个音轨—source_task_id, source_track, source_audio_id, infill_start_s, infill_end_s, full_lyrics, tags, title, negativeTags
suno-cover-generateaudio每个音轨—source_task_id
lyria-3-5

Google Lyria 3.5 — 根据风格描述生成音乐,可选择是否添加歌词;支持控制 BPM 和曲目标题。按曲目计费。

变体(addressAs)
参数类型描述
lyria-3-5:whitewhiteWhite · Available
lyria-3-5:flowflowFlow · 可用
opts 参数
参数类型描述
lyricstext歌词(可选)
titletext音轨标题
bpmnumberBPM(节奏)
seedtextSeed(可选)
  • variants: lyria-3-5:white, lyria-3-5:flow

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"lyria-3-5","prompt":"Calm electronic music without vocals, 90 BPM."}'
elevenlabs-music

ElevenLabs Music v2.5 — 根据描述生成音乐:包括流派、情绪、乐器,以及是否包含人声。曲长为 10 秒至 3 分钟。价格取决于曲长。

变体(addressAs)
参数类型描述
elevenlabs-musicflowFlow · 可用
opts 参数
参数类型描述
durationselect音轨时长(秒)· 默认值:30 · 选项:10, 30, 60, 120, 180
force_instrumentalbool仅器乐 · 默认值:false
output_formatselect格式 · 默认值:mp3 · 选项:mp3、mp3_hq
seednumberSeed(可选)
  • 变体:elevenlabs-music
  • duration: 10 | 30 | 60 | 120 | 180
  • output_format: mp3 | mp3_hq

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"elevenlabs-music","prompt":"Calm electronic music without vocals, 90 BPM.","opts":{"duration":"30","output_format":"mp3"}}'
acestep-v15-xl-sft

通过 Dune 提供的 ACE-Step v1.5 XL SFT。根据提示词和歌词生成 30–500 秒的音乐。

变体(addressAs)
参数类型描述
acestep-v15-xl-sftduneDune · 可用
opts 参数
参数类型描述
lyricstext歌词
durationnumber时长(30–500 秒)· 默认值:30
instrumentalbool器乐
language_codetext语言
  • variants: acestep-v15-xl-sft

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"acestep-v15-xl-sft","prompt":"Calm electronic music without vocals, 90 BPM.","opts":{"duration":"30"}}'
suno

Suno:根据描述生成音乐。版本包括 V6、V6 Wild、V6 Mini。支持人声或器乐。

变体(addressAs)
参数类型描述
sunopixelPixel · 可用
suno:whitewhiteWhite · Available
suno:amberamber琥珀色 · 不可用:所选媒体路由在这些参数下不可用。
opts 参数
参数类型描述
versionselect版本 · 默认:v6 · 选项:v6、v6-wild、v6-mini
instrumentalbool纯器乐(无 vocals) · 默认值:false
styletext风格 / 类型(最多 1000 个字符)
titletext音轨标题(最多 80 个字符)
negativeTagstext排除风格(例如“重金属、劲爆鼓点”)
vocalGenderselect声音性别 · 选项:、m、f
styleWeighttext风格遵循度(0–1)
weirdnessConstrainttextCreativity / deviation (0–1)
audioWeighttext音频特征平衡(0–1)
personaIdtextPersona ID / Voice ID(可选)
personaModelselectPersona 类型 · 选项:、style_persona、voice_persona
durationnumber时长(秒,10–360)
  • variants: suno, suno:white, suno:amber
  • version: v6 | v6-wild | v6-mini (affects price)
  • vocalGender: | m | f
  • personaModel: | style_persona | voice_persona

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno","prompt":"Calm electronic music without vocals, 90 BPM.","opts":{"version":"v6","vocalGender":"","personaModel":""}}'
suno-mashup

Suno Mashup:将恰好两条上传的音轨混音为一条。每次请求按固定价格计费。

opts 参数
参数类型描述
upload_urlsreq.array两个音频文件(恰好 2 个 URL)
versionselect版本 · 默认:v6 · 选项:v6、v6-wild、v6-mini
custom_modebool高级模式(自定义样式和标题)· 默认值:false
instrumentalbool纯器乐(无 vocals) · 默认值:false
styletext风格 / 类型
titletext音轨标题(最多 80 个字符)
vocalGenderselect声音性别 · 选项:、m、f
styleWeightnumber风格遵循度(0–1)
weirdnessConstraintnumberCreativity / deviation (0–1)
audioWeightnumber音频特征平衡(0–1)
  • 层级:每次请求
  • version: v6 | v6-wild | v6-mini
  • vocalGender: | m | f

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-mashup","prompt":"Calm electronic music without vocals, 90 BPM.","opts":{"upload_urls":["https://example.com/input.jpg"],"version":"v6","vocalGender":""}}'
suno-add-instrumental

Suno add-instrumental:为上传的人声轨道添加器乐部分。每次请求按固定价格计费。

opts 参数
参数类型描述
upload_urlreq.text音频(URL)
versionselect版本 · 默认:v6 · 选项:v6、v6-wild、v6-mini
titletext音轨标题(最多 80 个字符)
tagstext风格标签(例如 "relaxing, piano")
negativeTagstext排除风格
vocalGenderselect声音性别 · 选项:、m、f
styleWeightnumber风格遵循度(0–1)
weirdnessConstraintnumberCreativity / deviation (0–1)
audioWeightnumber音频特征平衡(0–1)
  • 层级:每次请求
  • version: v6 | v6-wild | v6-mini
  • vocalGender: | m | f

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-add-instrumental","prompt":"Calm electronic music without vocals, 90 BPM.","opts":{"version":"v6","vocalGender":""}}'
suno-add-vocals

Suno add-vocals:为上传的器乐轨道添加人声部分。每次请求按固定价格计费。

opts 参数
参数类型描述
upload_urlreq.text音频(URL)
versionselect版本 · 默认:v6 · 选项:v6、v6-wild、v6-mini
titlereq.text音轨标题(最多 80 个字符)
stylereq.text风格 / 类型
negativeTagsreq.text排除风格
vocalGenderselect声音性别 · 选项:、m、f
styleWeightnumber风格遵循度(0–1)
weirdnessConstraintnumberCreativity / deviation (0–1)
audioWeightnumber音频特征平衡(0–1)
  • 层级:每次请求
  • version: v6 | v6-wild | v6-mini
  • vocalGender: | m | f

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-add-vocals","prompt":"Calm electronic music without vocals, 90 BPM.","opts":{"version":"v6","vocalGender":""}}'
suno-cover-audio

Suno — 上传的音频文件或之前某个 Suno 音轨的翻唱版本。来源:upload_url,或使用 source_task_id(你的 Suno 作业 ID)和 source_track(音轨编号,0 或 1);二选一。每次请求统一价格。

opts 参数
参数类型描述
upload_urltext音频(URL),或提供 source_task_id
source_task_idtext你的 Suno 作业 ID(jobId),而不是 upload_url
source_tracknumber结果中的音轨编号(0、1、……)
versionselect版本 · 默认:v6 · 选项:v6、v6-wild、v6-mini
custom_modebool高级模式(自定义样式和标题)· 默认值:false
instrumentalbool纯器乐(无 vocals) · 默认值:false
styletext风格 / 类型
titletext音轨标题(最多 80 个字符)
negativeTagstext排除风格
vocalGenderselect声音性别 · 选项:、m、f
styleWeightnumber风格遵循度(0–1)
weirdnessConstraintnumberCreativity / deviation (0–1)
audioWeightnumber音频特征平衡(0–1)
personaIdtextPersona ID(可选)
  • 层级:每次请求
  • version: v6 | v6-wild | v6-mini
  • vocalGender: | m | f

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-cover-audio","prompt":"Calm electronic music without vocals, 90 BPM.","opts":{"version":"v6","vocalGender":""}}'
suno-extend-audio

Suno upload-and-extend-audio:延续上传的音频文件。每次请求按固定价格计费。

opts 参数
参数类型描述
upload_urlreq.text音频(URL)
versionselect版本 · 默认:v6 · 选项:v6、v6-wild、v6-mini
continue_atnumber从第二个继续
instrumentalbool纯器乐(无 vocals) · 默认值:false
styletext风格 / 类型
titletext音轨标题(最多 80 个字符)
negativeTagstext排除风格
vocalGenderselect声音性别 · 选项:、m、f
styleWeightnumber风格遵循度(0–1)
weirdnessConstraintnumberCreativity / deviation (0–1)
audioWeightnumber音频特征平衡(0–1)
personaIdtextPersona ID(可选)
  • 层级:每次请求
  • version: v6 | v6-wild | v6-mini
  • vocalGender: | m | f

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-extend-audio","prompt":"Calm electronic music without vocals, 90 BPM.","opts":{"version":"v6","vocalGender":""}}'
suno-replace-section

Suno Replace Music Section:替换之前生成的 Suno 内容中的一个时间范围。需要该内容的 task id 和 audio id。每次请求按固定价格计费。

opts 参数
参数类型描述
source_task_idreq.text你的 Suno 作业 ID(jobId)
source_tracknumber结果中的音轨编号(0、1、……)
source_audio_idtext源音轨 id(audioId)
infill_start_sreq.number分段开始时间(秒)
infill_end_sreq.number分段结束时间(秒)
full_lyricstext包括替换部分在内的完整歌词
tagstext风格标签
titletext音轨标题(最多 80 个字符)
negativeTagstext排除风格
  • 层级:每次请求

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-replace-section","prompt":"Calm electronic music without vocals, 90 BPM.","opts":{"infill_start_s":1,"infill_end_s":1}}'
suno-cover-generate

Suno Cover Generate:为之前生成的 Suno 内容制作翻唱。Pixel 将此路由发布为零积分,因此价格由所有者决定(2026-08-31),而不是根据成本利润率确定。

opts 参数
参数类型描述
source_task_idreq.text你的 Suno 作业 ID(jobId)
  • 层级:每次请求

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"suno-cover-generate","prompt":"Calm electronic music without vocals, 90 BPM."}'

3D models (7)

model输入单位变体opts 参数
hunyuan3dimage每个模型hunyuan3d:flowimage_url, textured_mesh, seed, guidance_scale, num_inference_steps, octree_resolution
hunyuan3d-2image每个模型—image_url
hunyuan3d-2mvimage每个模型—front_image, back_image, left_image, right_image
tripotext, image每个模型tripo:flowmode, texture, image_url, pbr
meshytext, image每个模型meshy:flowmode, image_url, should_texture, enable_pbr
rodintext, image每个模型rodin:flowmode, endpoint_speed, tier, format, image_url, input_urls, material
trellisimage每个模型trellis:flowimage_url, seed, ss_guidance_strength, ss_sampling_steps, slat_guidance_strength, slat_sampling_steps, mesh_simplify, texture_size
hunyuan3d

Hunyuan3D v2(Tencent)— 图像 → 3D 网格(.glb)。

变体(addressAs)
参数类型描述
hunyuan3d:flowflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
textured_meshbool带纹理 · 默认值:false
seednumberSeed
guidance_scalenumberGuidance (0–20)
num_inference_stepsnumberSteps (1–50)
octree_resolutionnumber八叉树分辨率(1–1024)
  • variants: hunyuan3d:flow

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"hunyuan3d","opts":{"image_url":"https://example.com/input.jpg"}}'
hunyuan3d-2仅目录

Hunyuan3D-2(Canvas):预计 GPU 时间价格;当前无法生成。

opts 参数
参数类型描述
image_urlreq.text图像(URL)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、displayOnly、generationAvailable: false

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"hunyuan3d-2","opts":{"image_url":"https://example.com/input.jpg"}}'
hunyuan3d-2mv仅目录

Hunyuan3D-2mv(Canvas):预计 GPU 时间价格;当前无法生成。

opts 参数
参数类型描述
front_imagereq.text正面视图
back_imagetext背面视图
left_imagetext左视图
right_imagetext右侧视图

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、displayOnly、generationAvailable: false

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"hunyuan3d-2mv","opts":{"front_image":"https://example.com/input.jpg"}}'
tripo

Tripo3D v2.5 — ultra-fast 3D generation. Image-to-3D and text-to-3D. Output .glb.

变体(addressAs)
参数类型描述
tripo:flowflowFlow · 可用
opts 参数
参数类型描述
modeselect模式 · 默认值:image-to-3d · 选项:image-to-3d、text-to-3d
textureselect纹理 · 默认值:standard · 选项:no、standard、HD
image_urltext图像(URL,用于图像转 3D 模式)
pbrboolPBR 材质 · 默认值:false
  • variants: tripo:flow
  • mode: image-to-3d | text-to-3d
  • texture: no | standard | HD (affects price)

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"tripo","prompt":"A minimalist chair in light wood.","opts":{"mode":"image-to-3d","texture":"standard"}}'
meshy

Meshy v6:适合游戏开发的最佳纹理。图像/文本转 3D。输出 .glb、.usdz、.fbx。

变体(addressAs)
参数类型描述
meshy:flowflowFlow · 可用
opts 参数
参数类型描述
modeselect模式 · 默认值:image-to-3d · 选项:image-to-3d、text-to-3d
image_urltext图像(URL,用于图像转 3D 模式)
should_texturebool生成纹理 · 默认值:true
enable_pbrboolPBR 材质 · 默认值:false
  • variants: meshy:flow
  • mode: image-to-3d | text-to-3d

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"meshy","prompt":"A minimalist chair in light wood.","opts":{"mode":"image-to-3d"}}'
rodin

Rodin(Hyper3D)v2.5 —— 复杂角色,高细节。输出 .glb/.usdz/.fbx/.obj/.stl。最多 5 张图像。

变体(addressAs)
参数类型描述
rodin:flowflowFlow · 可用
opts 参数
参数类型描述
modeselect模式 · 默认值:image-to-3d · 选项:image-to-3d、text-to-3d
endpoint_speedselect生成路由 · 默认值:standard · 选项:fast、standard
tierselect网格细节(Rodin 层级) · 默认值:Gen-2.5-High · 选项:Gen-2.5-Extreme-Low、Gen-2.5-High、Gen-2.5-Extreme-High
formatselect格式 · 默认值:glb · 选项:glb、usdz、fbx、obj、stl
image_urltext图像(URL,用于图像转 3D 模式)
input_urlsarray图像(1–5 个 URL,多视图)
materialselect材质 · 默认值:PBR · 选项:PBR、Shaded、All、None
  • variants: rodin:flow
  • mode: image-to-3d | text-to-3d
  • endpoint_speed: fast | standard (affects price)
  • tier: Gen-2.5-Extreme-Low | Gen-2.5-High | Gen-2.5-Extreme-High (affects price)
  • format: glb | usdz | fbx | obj | stl
  • material: PBR | Shaded | All | None

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub、axes[]

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"rodin","prompt":"A minimalist chair in light wood.","opts":{"mode":"image-to-3d","endpoint_speed":"standard","tier":"Gen-2.5-High","format":"glb","input_urls":["https://example.com/input.jpg"],"material":"PBR"}}'
trellis

Trellis (Microsoft) — 从图像即时生成 3D。输出 .glb。

变体(addressAs)
参数类型描述
trellis:flowflowFlow · 可用
opts 参数
参数类型描述
image_urlreq.text图像(URL)
seednumberSeed
ss_guidance_strengthnumberSS 引导(0–10)
ss_sampling_stepsnumberSS 步数(1–50)
slat_guidance_strengthnumberSLAT 引导(0–10)
slat_sampling_stepsnumberSLAT 步数(1–50)
mesh_simplifynumber网格简化(0.9–0.98)
texture_sizeselect纹理大小 · 默认值:1024 · 选项:512、1024、2048
  • variants: trellis:flow
  • texture_size: 512 | 1024 | 2048

catalog 中的价格字段:fromRub、pricesAreEstimate、priceUnavailable、variants[].addressAs、variants[].priceRub

curl https://getvibeapi.com/api/media/generate \
  -H "Authorization: Bearer $GETVIBE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"trellis","opts":{"image_url":"https://example.com/input.jpg","texture_size":"1024"}}'
常规生成流程、任务轮询和结果文件位于 Media generation 部分。Webhook、错误和计费位于 Webhooks and job lifecycle。
GetVibe API 媒体目录 — Seedance、Wan、Suno、TTS、3D | GetVibe API