Provider
Defaultdefault
kimi-k3Minimum charge per request — $0.0000143 (supplier rule): short requests are billed no less than this amount
Input $/1M
0.1758
Output $/1M
0.8786
Context
1M / 32K
Availability
—
TTFT
—
Speed
—
Cache hit
—
Caching
yes
GetVibeAPITheme
Language
Text
Kimi K3 API — how to connect and pay
A high-performance model tuned for a broad range of tasks.
Input ($/1M)
0.1758
Output ($/1M)
0.8786
Cache: read / write ($/1M)
0.01758 / 0.1758
Context
1M
Prices in $, updated October 25, 2026.
Performance
Last 7 days: responses / client requests that reached a supplier. Retries of one request count once; input checks, disabled routes, and client cancellations do not lower model quality.
Availability
60.63%
client requests over 7 days
Average latency
11.00 s
to full response
TTFT
6.02 s
to first token
Speed
63 tok/s
generation speed, tok/s (median)
Strengths
- OpenAI-compatible access
- Single key for all models
Good for
- Chat assistants
- Automation
- Text processing
- Product integration
Capabilities
- Context
- 1M tokens
- Max output
- 32K tokens
- Input
- text, image
- Tool calling
- supported
- JSON mode
- supported
- Vision
- yes
- Reasoning
- thinking — Reasoning tokens are billed at the output rate
Kimi K3 from Moonshot AI; supports reasoning and tools.
Providers
Comparison of public providers for this model: price, availability, TTFT, speed and cache. No measured metrics for these providers yet — showing «—», no invented numbers.
| Provider | Input $/1M | Output $/1M | Context | Availability | TTFT | Speed | Cache hit | Caching |
|---|---|---|---|---|---|---|---|---|
Defaultdefault kimi-k3Minimum charge per request — $0.0000143 (supplier rule): short requests are billed no less than this amount | 0.1758 | 0.8786 | 1M / 32K | — | — | — | — | yes |
Offers
One line per price level: our input / output price per 1M tokens, the discount off the official list price, and how many connected merchant subscription accounts are offering there.
:qwencloudqwencloud
Typical price you actually pay: $0.1046 / $0.5228 per 1M tokens
Based on 219 paid requests, last 7 days
$0.8572 / $4.29
$1.03 / $5.14
$1.07 / $5.36
$2.14 / $10.71
official $3 / $15
:qwencloud-alibaba-cnqwencloud-alibaba-cn
Typical price you actually pay: $0.4638 / $2.32 per 1M tokens
Based on 103 paid requests, last 24 hours
$0.4286 / $2.14
$0.5015 / $2.51
$0.5143 / $2.57
$2.14 / $10.71
official $3 / $15
:codebuddycodebuddy
Typical price you actually pay: $0.1581 / $0.7901 per 1M tokens
Based on 203 paid requests, last 24 hours
$0.2358 / $1.18
$0.2443 / $1.22
$0.2572 / $1.29
$0.2786 / $1.39
$0.2829 / $1.41
$0.3 / $1.5
official $3 / $15
:codebuddy-cncodebuddy-cn
Typical price you actually pay: $0.1285 / $0.6421 per 1M tokens
Based on 1,820 paid requests, last 24 hours
$0.1758 / $0.8786
$0.1843 / $0.9215
$0.1929 / $0.9643
$0.1972 / $0.9858
$0.2315 / $1.16
$0.2443 / $1.22
official $3 / $15
:kiro-poolkiro-pool
$1.71 / $8.55
$2.14 / $10.71
official $3 / $15
market only
clinepass
No offers right now
market only
opencode
No offers right now
OpenAI-compatible endpoint
POST https://getvibeapi.com/v1/chat/completionsAuthorization
Pass your API key in the Bearer header. Keys start with sk-pv-.
Authorization: Bearer $GETVIBE_API_KEYResponse format
Responses follow the Chat Completions contract: the text is in choices[0].message.content, token usage is in usage.
First request
All examples use the public model address kimi-k3.
curl https://getvibeapi.com/v1/chat/completions \
-H "Authorization: Bearer $GETVIBE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "kimi-k3",
"messages": [
{
"role": "user",
"content": "Explain quantum computing in simple terms"
}
],
"stream": false
}'Chat Completions
Fields are generated from the service's live OpenAPI schema.
Field
modelType
string
Required
yes
Description
Public model ID in vendor/model form (OpenRouter-style), e.g. anthropic/claude-opus-5. The bare claude-opus-5 spelling is accepted as an input alias and bills identically. model:alias pins a channel, e.g. openai/gpt-5.6-sol:yellow. GET /v1/models lists one ID per model by default; ?variants=1 returns the full list.
Field
messagesType
array
Required
yes
Description
No description
Field
streamType
boolean
Required
no
Description
Stream the response as server-sent events (SSE)Default: false
Field
max_tokensType
integer
Required
no
Description
Maximum number of tokens in the response
Field
temperatureType
number
Required
no
Description
Sampling randomness, 0–2Default: 1
Field
top_pType
number
Required
no
Description
No description
Field
reasoning_effortType
string
Required
no
Description
Reasoning level for Chat Completions. The supported values depend on the model.none · minimal · low · medium · high · xhigh · max
Field
toolsType
array
Required
no
Description
List of tools (function calling)
Field
tool_choiceType
string
Required
no
Description
auto, none, required or a specific tool
Field
response_formatType
object
Required
no
Description
Force JSON output
| Field | Type | Required | Description |
|---|---|---|---|
model | string | yes | Public model ID in vendor/model form (OpenRouter-style), e.g. anthropic/claude-opus-5. The bare claude-opus-5 spelling is accepted as an input alias and bills identically. model:alias pins a channel, e.g. openai/gpt-5.6-sol:yellow. GET /v1/models lists one ID per model by default; ?variants=1 returns the full list. |
messages | array | yes | No description |
stream | boolean | no | Stream the response as server-sent events (SSE)Default: false |
max_tokens | integer | no | Maximum number of tokens in the response |
temperature | number | no | Sampling randomness, 0–2Default: 1 |
top_p | number | no | No description |
reasoning_effort | string | no | Reasoning level for Chat Completions. The supported values depend on the model.none · minimal · low · medium · high · xhigh · max |
tools | array | no | List of tools (function calling) |
tool_choice | string | no | auto, none, required or a specific tool |
response_format | object | no | Force JSON output |
Kimi K3 API
A high-performance model tuned for a broad range of tasks.
Install
pip install openaiQuick start
from openai import OpenAI
client = OpenAI(
api_key="sk-pv-...",
base_url="https://getvibeapi.com/v1",
)
response = client.chat.completions.create(
model="kimi-k3",
messages=[{"role": "user", "content": "Explain quantum computing in simple terms"}],
)
print(response.choices[0].message.content)Configuration
- base_url
https://getvibeapi.com/v1- model
kimi-k3- api_key
- GetVibe key from the dashboard
Full documentation: getvibeapi.com/docs.
Key facts
The live catalog contains 74 text models.
Kimi K3: 0.1758 $ input and 0.8786 $ output per 1M tokens.
FAQ about the Kimi K3 API
How much does Kimi K3 cost?
The price of Kimi K3: $0.1758 per 1M input · $0.8786 per 1M output tokens. Pay only for the tokens you use — no subscriptions or flat fees.
Is it compatible with the OpenAI SDK?
Yes, the API is fully OpenAI-compatible. Just change base_url to https://getvibeapi.com/v1 and set your key — the rest of the code stays the same.