Here is what is new at GetVibe API: a redesigned dashboard, channel ratings, transparent billing, new models and higher rate limits.
New dashboard
A cleaner, faster cabinet with a proper mobile version. The model providers window now shows each channel's share of traffic, capacity and speed — pick or exclude a channel in one click.
Rate channels
Like or dislike any channel right in the providers window. We read every rating and use them to decide which suppliers stay.
Transparent billing
Every request in your log now shows the exact price tier it was filled at, and the lines add up to what you were charged. Channel names in your logs are now human-readable.
New text models
Live prices:
- Claude Haiku 5.5 — $0.0267 input · $0.1334 output per 1M tokens
- NVIDIA Nemotron 3 Super 120B — $0.1417 input · $0.6667 output per 1M tokens
- NVIDIA Nemotron 3.5 Lightning — $0.10 input · $0.2667 output per 1M tokens
- Llama 4 Maverick — $0.3334 input · $1.3334 output per 1M tokens
Images
Nano Banana 2.1 (up to 10 references, up to 4K), Seedream 5.0 Flash, FLUX.2 Max, Qwen Image 2.1.
Voice and avatars
- MiMo TTS (voice cloning and voice design) and Inworld TTS 1.5;
- speech recognition with Nemotron ASR, Qwen3 ASR and Voxtral, with speaker diarization;
- ACE-Step 1.5 music;
- HeyGen Avatar 4 and P-Video Avatar talking avatars.
Higher rate limits
From a balance of 2 USD: 6,600 requests/min and 300 concurrent; from 12 USD: 44,000/min and 2,000 concurrent. Even with a zero balance: 300 requests/min and 10 concurrent.
Everything above is available now in your getvibeapi.com dashboard.