LLM API Gateway
One asynchronous chat API for Anthropic and Fireworks, with consistent responses and streaming.
- Revocable keys, provider permissions, and atomic rate limits.
- Per-caller response caching; streaming requests are never cached.
- Python
- FastAPI
- HTTPX
- SQLite
AWS · CI/CD · latency and cost tracking
View on GitHub (opens in a new tab)Scoped keys. Consistent streaming. Measured usage.

