About this offer
UnoRouter is a failover-first model gateway for engineering teams and AI agents that need model calls to keep succeeding when one provider has an outage or hits a rate limit. It exposes a single OpenAI-compatible endpoint and key; each request goes to a healthy upstream and is retried on another provider if the first one fails. Integration is a base URL change. The official OpenAI SDKs, Claude Code, Cline, OpenCode and other AI clients work unchanged, and moving to a different model means changing the model name in the request. The gateway, web app and sync tooling are open source. The hosted chat is local-first, so conversation history stays in the user's browser. Free models are available on the free tier, and usage beyond it is billed pay-as-you-go.
Details from the publisher
- Rate and usage limits
- Free tier limits, then usage based billing
- Supported operations
- Chat completions, model listing, OpenAI compatible endpoints
- Documentation URL
- https://unorouter.com/docs
- SDK languages
- Any OpenAI SDK (Python, JavaScript, Go and others)
- Authentication
- API key (Bearer token)
LLM gateway · OpenAI compatible API · model routing · failover · open source