Documentation
Vouch Compute API docs
Vouch Compute speaks the OpenAI API. If your code already calls OpenAI for transcription or embeddings, you change the base URL and the API key, and nothing else.
Distributed processing. Jobs run on a network of vetted consumer GPUs and Macs owned by independent contributors. Anyone running a job can see its input. Do not send personal or sensitive data. Read the data policy.
Start here
- Quickstart: first transcription and first embedding in under 10 minutes.
- Batch API: long files and archives. This is the cheapest path.
- API reference: every public endpoint, generated from the OpenAPI spec.
What we serve
| Workload | Models | Endpoint |
|---|---|---|
| Transcription, async batch | whisper-large-v3-turbo, whisper-large-v3, whisper-small | POST /v1/audio/transcriptions/jobs |
| Transcription, sync (25 MB max) | same | POST /v1/audio/transcriptions |
| Embeddings | bge-m3, bge-small-en-v1.5 | POST /v1/embeddings |
Chat completions are not offered. POST /v1/chat/completions returns model_not_found for every model.