A free MIT AI gateway providing a unified endpoint to 290+ providers and 500+ models with quota-aware auto-fallback and token compression.
- v3.8.53
- Published at
- 7d ago
- 87,103
- Stars
- 151,523
- Download
- Synced at
- 12m ago
Latest release
v3.8.53Nov 6, 2026, 10:53 AMDownloads slow? Switch the source anytime above
README
Project Overview
OmniRoute is a free MIT AI gateway. It provides a unified endpoint to 290+ providers (90+ free) and 500+ models. Features quota-aware auto-fallback and RTK+Caveman compression saving 15-95% tokens.
Key Features
- Unified endpoint to 290+ AI providers
- 500+ models supported (Kimi, Claude, GPT, OpenAI, Gemini, GLM, DeepSeek, etc.)
- Quota-aware auto-fallback
- RTK+Caveman compression saving 15-95% tokens
- Compatible with Claude Code, Codex, Cursor, OpenCode, Cline, Copilot, and more
- MCP/A2A support
- Desktop and PWA versions
- ~1.53 billion free tokens per month
Target Audience
- Developers needing access to multiple AI models
- Users wanting to reduce API call costs
- Teams building AI applications
Quick Start
- Visit the OmniRoute repository for installation guides
- Install via Docker, source, or pnpm
- Zero-configuration out of the box
- Check the dashboard for free tier usage
Core Advantages
- Completely free with MIT license
- Aggregates 90+ free providers offering substantial free tokens
- Smart fallback and compression to reduce costs
- Built by 500+ contributors
System Requirements
- OS: Cross-platform (Desktop and PWA)
- Architectures: x86_64, arm64
- Additional Dependencies: Varies by installation method
Notes
- Free tier quotas fluctuate with provider policy changes
- Free tier quotas are re-audited every two weeks
Release history
Sort
Related projects
ollama
ollama/ollama
Running open models locally no longer needs the cloud — ollama compresses downloading, running, and serving into a few commands, offering a CLI, a REST API, and official Python / JS libraries across macOS, Windows, Linux, and Docker.
v0.40.17-rc0 Go
Author updated 2d ago
llama.cpp
ggml-org/llama.cpp
A C/C++ implementation for LLM and VLM inference with minimal setup and state-of-the-art performance, supporting multiple model formats with CLI, server, and built-in Web UI.
b12362 C++
Author updated 13m ago
vllm
vllm-project/vllm
vLLM is a high-throughput, memory-efficient LLM inference and serving engine (Python). Built on PagedAttention and continuous batching, it serves 200+ model architectures behind an OpenAI-compatible API across many hardware backends.
v0.33.0rc3 Python
Author updated 11d ago
freellmapi
tashfeenahmed/freellmapi
A unified API gateway aggregating 34 free LLM providers and 635 free model endpoints, offering approximately 7.4 billion free tokens per month with smart routing, automatic failover, and encrypted key storage.
v0.16.28 TypeScript
Author updated 7h ago
llmfit
AlexsJones/llmfit
A command-line tool that recommends the best local LLMs for your hardware configuration, supporting multi-GPU, MoE architectures, and multiple quantization formats.
v1.1.20 Rust
Author updated 10d ago
magpie
yetone/magpie
Magpie is a locally running AI gateway that lets coding agents such as Claude Code, Codex, Gemini CLI and OpenCode use models like Kimi, DeepSeek, GLM and ChatGPT, with one-click menu bar switching and automatic failover to the next account when quota runs out.
v0.1.4082 Go
Author updated 54m ago