Devtoolwatch BoardOmniRoute review
OmniRoute is an open-source AI gateway that routes requests across many model providers through one OpenAI-compatible endpoint, with automatic fallback.
TL;DR
OmniRoute is a self-hosted AI gateway that puts your connected model providers behind one local endpoint. It suits developers who use coding clients like Claude Code, Codex, Cursor or Cline and want quota-aware fallback across providers. It also offers auto-routing combos, multiple routing strategies and context compression.
- 83.2kGitHub stars
- +2.6knew stars, week of Oct 25
- ~106kdownloads a week, +55% vs 4-week avg
- #5in Inference and training this week
OmniRoute at a glance
- Category
- Inference and training
- Maintained by
- Cheaper Inference (organization, @cheaper-inference on GitHub)
- License
- MIT
- Language
- TypeScript, JavaScript, Shell
- Repository
- github.com/cheaper-inference/OmniRoute
- Website
- omniroute.online
- npm package
- omniroute 3.8.52
- Latest release
- v3.8.52 on GitHub, Oct 19, 2026
- First published
- Feb 13, 2026
What OmniRoute does
- OmniRoute exposes one local endpoint at http://localhost:20128/v1 that accepts OpenAI, Claude, Gemini and Responses API requests.
- OmniRoute routes requests through combos, which are chains of models that move to the next eligible healthy model when quota runs out or a provider fails.
- OmniRoute offers zero-config auto models such as auto/coding, auto/fast and auto/cheap that build a virtual combo from your connected providers.
- OmniRoute provides routing strategies including priority, round-robin, cost-optimized, cache-optimized, fusion and pipeline, which you can mix per combo step.
- OmniRoute compresses eligible context with RTK and Caveman compression, which the project says saves 15–95% of tokens.
- OmniRoute supports MCP and A2A and runs as a desktop app or PWA.
- OmniRoute shows a free-tier dashboard at /dashboard/free-tiers that catalogs third-party free-tier quotas.
How it works
OmniRoute runs locally as a gateway, distributed on npm and Docker Hub. You connect your own provider accounts through OAuth, API keys, web cookies, local runtimes or keyless routes. Then you copy your OmniRoute key from Dashboard → Endpoints and point clients at http://localhost:20128/v1. A headless mode is available through serve --headless.
How to install and start using OmniRoute
From npm
npm install -g omnirouteRequirements
- Node.js >=22.22.2 <23 or >=24.0.0 <27 for the npm package.
- Your own provider accounts and credentials where required. OmniRoute does not grant tokens itself.
First steps
- Connect at least one provider account, because the auto model needs an eligible route and a fresh install may have none.
- Copy your OmniRoute key from Dashboard → Endpoints.
- Send a chat completion to http://localhost:20128/v1/chat/completions with the model set to auto and your key as a Bearer token.
- Point your coding client, such as Claude Code, Codex or Cursor, at the local /v1 endpoint.
Who OmniRoute is for
- Developers using Claude Code, Codex, Cursor, Cline, OpenCode or Copilot who want one endpoint for many providers.
- Developers combining third-party free tiers and subscriptions with quota-aware fallback.
- Teams wanting to self-host an OpenAI-compatible router with custom routing strategies.
Limitations to know
- OmniRoute free-tier capacity depends on third-party providers, which control their own limits, availability and terms.
- OmniRoute's auto model needs an eligible route, and keyless providers may reject third-party clients.
- OmniRoute excludes providers marked tos: avoid, including Kiro, from automatic routing by default, even when you connect an account.
- OmniRoute's npm package for 3.8.47 crashed on boot and was deprecated, so releases have shipped packaging bugs.
Is OmniRoute growing? Usage data
In the week of Oct 25–31, 2026, OmniRoute ranked #32 on the Devtoolwatch Board and #5 in Inference and training, with a momentum score of 82 out of 100. Its main package had about 106k downloads that week, +55% against its own 4-week average.
Show the weekly numbers
| Week of | New stars |
|---|---|
| Aug 9, 2026 | 5.1k |
| Aug 16, 2026 | 4.7k |
| Aug 23, 2026 | 4.9k |
| Aug 30, 2026 | 3.5k |
| Sep 6, 2026 | 3.8k |
| Sep 13, 2026 | 3k |
| Sep 20, 2026 | 2.3k |
| Sep 27, 2026 | 2.3k |
| Oct 4, 2026 | 2.4k |
| Oct 11, 2026 | 2.9k |
| Oct 18, 2026 | 3.2k |
| Oct 25, 2026 | 2.6k |
The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.
OmniRoute alternatives
Other LLM inference and training tools on the Board in the week of Oct 25–31, 2026, by momentum:
| Rank | Tool | Stars | Downloads/wk |
|---|---|---|---|
| #1 | dsh-our-free-modeldsh-our-free-model is a dsh plugin that provides DeepSeek V4.1 Flash, Kimi K3 and other models without login, sign-up or API key. | 54.3k | – |
| #2 | magpiemagpie is a local gateway and app that lets coding agents like Claude Code, Codex and Gemini CLI use any provider's model. | 26.9k | – |
| #3 | StrataStrata is a local inference engine and app that runs large open models like Qwen3.8-Flash-Next on consumer hardware with… | 37.1k | – |
| #4 | edge0edge0 is an open-source streaming inference framework that runs large sparse mixture-of-experts models on consumer hardware by streaming… | 8.9k | – |
| #6 | CLIProxyAPICLIProxyAPI is a Go proxy server that exposes CLI coding accounts like Codex, Claude Code and Gemini CLI through OpenAI, Gemini and… | 57.1k | – |
| #7 | codex-chatgpt-webcodex-chatgpt-web is a launcher that lets Codex use the ChatGPT Web models on your account, including Pro, through browser automation. | 15.3k | – |
Recent OmniRoute releases
- v3.8.52, Oct 19, 2026
- v3.8.51, Sep 30, 2026
- v3.8.50, Aug 26, 2026
- radar-export-latest, Aug 20, 2026
- v3.8.49, Jul 30, 2026
OmniRoute FAQ
Is OmniRoute free and open source?
Yes. OmniRoute's source code is on GitHub (cheaper-inference/OmniRoute) under the MIT license, so you can use it for free.
How do I install OmniRoute?
Run npm install -g omniroute. It needs Node.js >=22.22.2 <23 || >=24.0.0 <27.
Is OmniRoute still maintained?
Yes. The cheaper-inference/OmniRoute repository was last updated on Oct 22, 2026, and the latest release, v3.8.52 on GitHub, came out on Oct 19, 2026.
How popular is OmniRoute?
OmniRoute has 83.2k GitHub stars as of Nov 1, 2026. It gained 2.6k new stars in the week of Oct 25–31, 2026, against a 12-week average of 3.4k a week. Its main package had about 106k downloads that week, +55% against its own 4-week average. That made it #5 among LLM inference and training tools on the Devtoolwatch Board.
OmniRoute vs LiteLLM: how do they compare?
OmniRoute and LiteLLM are both tracked by Devtoolwatch. OmniRoute has 83.2k GitHub stars and LiteLLM has 62.5k. In the week of Oct 25–31, 2026, OmniRoute's main package had 106k downloads and LiteLLM's had 23.1M. LiteLLM is an open source AI gateway and Python SDK that calls 100+ LLM providers through one OpenAI-format interface. LiteLLM review.
OmniRoute vs OpenCode: how do they compare?
OmniRoute and OpenCode are both tracked by Devtoolwatch. OmniRoute has 83.2k GitHub stars and OpenCode has 223.6k. OpenCode is an open source AI coding agent that helps you write and run code in a terminal, desktop app, or IDE extension. OpenCode review.
What are the best alternatives to OmniRoute?
Other LLM inference and training tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: dsh-our-free-model (#1), magpie (#2), Strata (#3), edge0 (#4), CLIProxyAPI (#6). The alternatives table above has their numbers.
Is OmniRoute safe to use?
OmniRoute lists encrypted keys and a circuit breaker among its operational controls. By default, it excludes providers marked tos: avoid from automatic routing. The README advises reading the Free Tiers Guide before choosing a provider, and never committing the generated local .env file.
What happens when a provider runs out of quota?
OmniRoute moves to the next eligible healthy model in the combo when quota runs out, a provider fails or costs spike. The auto/subscription mode only uses subscription quota and fails closed with no billable fallback.
Does OmniRoute give me free tokens?
No. OmniRoute does not grant tokens. Its free-tier figure is an estimate across third-party free tiers you sign up for yourself. Each provider controls its own limits and terms.
Sources
OmniRoute README badge
Maintainers can show the tool's weekly Board rank in their README:
[](https://index.devtoolwatch.com/tools/diegosouzapw-omniroute/)How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].