Self-host with Docker · Yantr
OmniRoute
Free AI gateway routing 290+ providers through one OpenAI-compatible endpoint.
1
Ports
3
Notes
3
Use Cases
About
What OmniRoute does
OmniRoute is a free, self-hosted AI gateway that aggregates 290+ LLM providers (90+ with free tiers) behind a single OpenAI-compatible endpoint. It features auto-fallback across providers, token compression saving 15-95%, smart routing strategies, and works with Claude Code, Cursor, Codex, Cline, and Copilot out of the box.
Install Flow
How to self-host OmniRoute
OmniRoute can be deployed with standard Docker Compose using the Yantr app catalog. This page collects the basic information searchers usually need first: what OmniRoute does, which Docker image Yantr uses, default ports, deployment notes, and where to find the original compose template.
After deployment, Yantr lets you publish additional ports from the stack page using Docker-style syntax. Use 8080:8080 for a fixed host port or just 8080 to let Docker assign a random port.
Use Cases
What people run it for
- Route all your AI coding tools through one endpoint with automatic failover when a provider hits rate limits.
- Save up to 95% on token costs using built-in RTK and Caveman compression across free and paid LLM tiers.
- Run a local AI proxy that supports every major provider — OpenAI, Anthropic, Google, DeepSeek, Mistral, and 280+ more.
Technical Details
Ports, image, and service metadata
Ports
| Port | Protocol | Label |
|---|---|---|
| 20128 | HTTP | Dashboard |
Docker Image
diegosouzapw/omniroute:latest
Service Name
omniroute
Notes
Things to know before you deploy
- OmniRoute works out of the box with keyless free providers — no API keys needed to start.
- Add API keys for paid providers via the web dashboard after deployment.
- Use the 'auto' model name for smart routing across all connected providers.
Related Apps
More tools you might pair with OmniRoute
Page generated 2026-07-27T04:28:00.415Z