Polyrouter
polyrouter.app
- Free plan
- Web
- API
- Self-hosted
What Polyrouter is
Polyrouter is a free, open-source LLM gateway that places one endpoint in front of multiple model providers. Requests can be routed by explicit model selection, header rules, routing tiers, or optional automatic routing layers. Ordered fallback chains and provider-specific circuit breakers help continue requests when a provider degrades. Spend budgets can apply globally or per agent over daily, weekly, or monthly periods, with alerts or blocking at chosen thresholds. Polyrouter accepts OpenAI- and Anthropic-compatible APIs and translates between those protocols. Providers can use API keys, OAuth subscriptions, compatible custom endpoints, or self-hosted models including Ollama, LM Studio, and llama.cpp. Clients need a configurable base URL and API key. Logs contain metadata by default; prompt and response capture is optional and self-hosted. Credentials are encrypted at rest, and supplied URLs are checked against private and metadata address ranges. Self-hosting uses Docker Compose v2 with PostgreSQL 16 and Redis. The software is licensed under AGPL-3.0.
Who it suits
Polyrouter suits developers and teams managing several hosted or local language-model providers through a shared API endpoint. It also fits organizations wanting self-hosted logging, routing, and budget controls.
Good to know, and what to watch
- One endpoint for multiple providers
- Ordered fallbacks and circuit breakers
- Global or per-agent spending budgets
- OpenAI and Anthropic protocol translation
- Metadata-only logs by default
- Self-hosting requires Docker Compose v2
- Deployment also requires PostgreSQL 16 and Redis
- Cloud hosting is listed as coming soon
- AGPL-3.0 license applies
Plans
Every plan and priceSelf-hosted
Free plan
The verdict
Polyrouter centralizes provider routing, compatibility, fallback behavior, and spending controls. Its free self-hosted model gives teams control over deployment and optional prompt capture.
What the maker publishes
- Agent compatibility
- Any client that accepts a configurable base URL and API key can connect, including OpenAI- and Anthropic-compatible SDKs. source
- Budgets
- Spend budgets can apply globally or per agent over day, week, or month windows and can alert or block requests at a threshold. source
- Deployment
- Self-hosting requires Docker with Compose v2 and runs the SPA, API, and proxy alongside PostgreSQL 16 and Redis. source
- Fallbacks
- Ordered fallback chains and per-provider circuit breakers help requests continue when a provider degrades. source
- Integrations
- Provider options include API keys, OAuth subscriptions, compatible custom endpoints, and self-hosted local models including Ollama, LM Studio, and llama.cpp. source
- License
- Polyrouter is open source under the AGPL-3.0 license. source
- Maker
- Polyrouter is built and maintained by Anthony Izzo. source
- Managed hosting
- Polyrouter Cloud is described as coming soon, with a waitlist for early access. source
- Privacy
- Request logs are metadata-only by default, and prompt or response capture is an opt-in self-hosted feature that is off by default. source
- Product
- Polyrouter is an open-source LLM gateway that puts one endpoint in front of multiple model providers. source
- Protocols
- It supports OpenAI- and Anthropic-compatible APIs and translates requests between the two protocols. source
- Routing
- It supports explicit model selection, header rules, routing tiers, and optional automatic routing layers. source
- Security
- Provider and notification credentials are encrypted at rest, and user-supplied URLs are checked against private and metadata address ranges. source
- Support
- The maker directs bug reports, feature requests, and questions to GitHub. source
Checked against the maker's pages on 8 Oct 2026.