Blaick API¶
The Blaick API gives you a single, unified interface to many generative AI models — Anthropic Claude, OpenAI GPT, Groq, AWS Bedrock, GCP Vertex AI, and local models — behind one authentication scheme, one billing model, and one safety layer.
If you can make an HTTPS request, you can build on Blaick.
What you can do¶
- Chat completions — send messages and get model responses, streaming or all at once.
- Pick from many models — one request shape works across every provider. Or let the auto-router choose the cheapest model that can do the job.
- Keep conversations — server-side conversation state so you don't have to resend history.
- Track spend — every response reports token usage and cost; query balance and history any time.
- Build agents — create scheduled, tool-using agents with a per-user secret vault.
How requests work¶
Every request is an HTTPS call to https://api.blaick.ai/api/v1 with a Bearer token:
- Authentication is a Bearer token — either a JWT from login or an API key with a
blk_prefix. See Authentication. - Requests and responses are JSON, except streaming responses which use Server-Sent Events.
- Errors use standard HTTP status codes with a structured JSON body. See Errors & Rate Limits.
A note on tokens¶
Blaick bills in tokens, priced simply:
$1 = 1,000 tokens (1 US cent = 10 tokens)
Each response tells you exactly how many tokens it cost. New accounts start on a 14-day free trial. See Usage & Tokens for the full model.
Next steps¶
- :material-rocket-launch: Quickstart — your first response in five minutes.
- :material-key: Authentication — create an API key.
- :material-message-processing: Chat Completions — the core endpoint, in depth.
Questions or issues? Reach us at support@blaick.ai.