cheaper-inference /OmniRoute
Never stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 900+ contributors
★ 86,100⑂ 13,075👁 490⑃ 13,063● 5,540▤ 283◈ 580
86,100
stars
+289
13,075
forks
490
watchers
13,063
prs
5,540
issues
283
releases
580
discussions
README
🚀 OmniRoute — The Free AI Gateway
Created by Diego Souza
💰 ~1.62B Tokens / Month Across Third-Party Free Tiers
Bring your own provider accounts. This is an estimated aggregate across separately eligible third-party free tiers, not a token grant from OmniRoute. Sign up, obtain credentials where required, and connect the providers you can use; each provider controls its limits, availability, and terms.
Stacking free tiers by hand is painful — dozens of SDKs, dozens of rate limits, and no idea how much you actually have. OmniRoute catalogs 465 free-tier entries across 34 recurring pool keys and computes the token headline from the 17 pools with a published positive monthly budget plus five per-model Groq caps, deduplicated by shared pool. Quotas that only open after a regional identity check (today: ModelScope) are shown apart, +~6M behind regional identity verification, and never summed into the headline. The result stays visible on the dashboard (/dashboard/free-tiers).
Animated summary of the live /dashboard/free-tiers page. Full methodology (pool dedupe, credit tiers, provider terms): docs/reference/FREE_TIERS.md.
These figures are re-audited every two weeks against the live catalog and move both ways — a provider ends a free tier and the number drops; a new one lands and it climbs. We publish what the catalog actually computes, never a rounded-up best case.
⭐ Star the repo if OMNIROUTE helped you save money and make your work easier.
💬 Join the community
👋 Follow the maintainer — get new providers, releases & tips first:
Questions, provider tips, roadmap & support → Discord · Telegram · WhatsApp 🌍 Global / 🇧🇷 Brasil / 📢 Official Channel / Portal
📈 The Gateway Keeps Growing
| | v3.8.50 | v3.8.51 | v3.8.54+ |
| ------------------------- | :-----------------------: | :---------: | :-------------------: |
| 🌐 Providers | 352 | 358 | 384 |
| 🧠 Unique chat model IDs | 1320 | 1374 | 1491 |
| 🖼️ Modality Bridge | 🆕 vision + audio + video | ✓ | ✓ |
| 📡 Radar free catalog | 🆕 opt-in | ✓ | ✓ |
| ⚖️ Quota-aware scheduling | 🆕 Quota-Share | ✓ | ✓ |
| 📊 Quota telemetry | 🆕 live | ✓ | ✓ |
| 🧰 Headless mode | — | — | 🆕 serve --headless |
| 🛤️ LTS rail infra | — | — | 🆕 release channels |
| 🔎 Model discovery API | — | — | 🆕 /v1/models/query |
→ Roadmap — riding the rail to v3.9.0 LTS
🧩 Available
!NPM Monthly
!Docker Pulls
!Electron Downloads
🚀 Start
🚀 Quick Start
📦 Install
🆓 Connect a provider
💡 Learn
💥 The PrTop issues
#17264🌙 nightly-compat: Node 24/26 failures on release/v3.8.5313 comments · 2026-10-20#16538fix(providers): Providers and models disappear after updating to 3.8.519 comments · 2026-10-13#16460fix(providers): Cloudflare Playground cannot load Playwright when sending an image9 comments · 2026-10-13#16454feat(backend): OTLP log-export destination, with destination runs isolated in a child process9 comments · 2026-10-13#18335fix(providers): Model import (sync-models) takes ~120s for real-discovery providers — silent timer-based wait in persist phase after fetch completes8 comments · 2026-10-27#17299fix(providers): codex failed toll call when JSON in arguments has Python regexp \{ JSON Error: Expecting ',' delimiter: line 1 column 41 (char 40)7 comments · 2026-10-20#16279docs(free-tier): re-verify Mistral's free plan (user report: free plan no longer activates API keys)7 comments · 2026-10-10#17196fix(api): /api/v1/providers/{provider}/models still returns stale Content-Length (regression of #14092) – dashboard model picker stays empty6 comments · 2026-10-19
Discussions
#19260Ollama Cloud free tier will soon be killed.0 comments · 2026-11-07#19195NEW FEATURE IDEA: Model Groups/Profiles/Tags1 comments · 2026-11-05#19079opencode-cache-manager: keep OpenCode's prompt cache warm, including through OmniRoute1 comments · 2026-11-05#19045Antigravity CLI as provider bannable ?2 comments · 2026-11-04#18810[RFC / Design Discussion]: Native iOS Companion Client for Remote OmniRoute Instances2 comments · 2026-11-03#18747Compression exclusion doesn't work (3.8.52)1 comments · 2026-11-02#18592Consolidate Kimi providers under kimi while preserving moonshot compatibility1 comments · 2026-10-30#18543Reasoning routing: effortMode=force with targetEffort=none does not remove effort fields (REASONING_ROUTING.md vs 3.8.52 behavior)2 comments · 2026-10-29
Latest release
v3.8.53 v3.8.53 2026-11-06OmniRoute v3.8.53 1,189 documented changes from 1,162 commits in the cycle, with 222 people contributing. Thank you all. 📄 Complete release notes (every entry, full description and credit): attached below as CHANGELOG-v3.8.53.md, and in [CHANGELOG.md → [3.8.53]](https://github.com/cheaper-inference/OmniRoute/blob/main/CHANGELOG.md). ⚠️ Known at release time: providers.apiRegionLabel (fi, ko, pl, pt, pt-BR, zh-CN, zh-TW) and providers.apiRegionInternational (fi) show their English text until they are retranslated in v3.8.54. 📊 Release by the numbers | | | | --- | ---: | | 👥 People who con