Free AI Proxy Gateway — OpenAI-Compatible LLM Proxy

Run your own free AI proxy — route requests across Groq, Gemini, Mistral, Cohere, OpenRouter, and 25+ providers through a single OpenAI-compatible endpoint.

🌐 See all proxy-ready providers at freellms.org

---

What is a Free AI Proxy Gateway?

An AI proxy gateway sits between your app and multiple LLM providers. You send one request format (OpenAI-compatible), the proxy routes it to whichever free provider you choose — Groq for speed, Gemini for long context, Mistral for code, OpenRouter for variety.

This repo documents how to build and use a free AI proxy using only free-tier LLM APIs. No paid keys, no usage fees.

Provider directory & base URLs: freellms.org

---

Supported Free Providers (Proxy-Ready)

Provider	Base URL	Free Tier	Proxy-Friendly
Groq	https://api.groq.com/openai/v1	30 RPM	✅
Google Gemini	https://generativelanguage.googleapis.com/v1beta	15 RPM	✅
Mistral AI	https://api.mistral.ai/v1	Free tier	✅
Cohere	https://api.cohere.com/v2	20 RPM	✅
OpenRouter	https://openrouter.ai/api/v1	1K RPD	✅
NVIDIA NIM	https://integrate.api.nvidia.com/v1	Free tier	✅
Cloudflare Workers AI	https://api.cloudflare.com/client/v4/accounts/{id}/ai/run	10K neurons/day	✅
Cerebras	https://api.cerebras.ai/v1	5 RPM	✅
SambaNova	https://api.sambanova.ai/v1	20 RPM	✅
LLM7.io	https://api.llm7.io/v1	30 RPM	✅

All base URLs + free API keys: freellms.org

---

Quick Start — Build a Free AI Proxy in 30 Seconds

1. Pick your upstream provider

export OPENAI_BASE_URL="https://api.groq.com/openai/v1"   # free, no credit card
export OPENAI_API_KEY="your-groq-key"                      # get at freellms.org

2. Point any tool at your proxy

Any client that accepts baseURL + apiKey is now a multi-provider proxy. Swap the base URL to switch providers — no code changes.

Python (OpenAI SDK as proxy client)

from openai import OpenAI

client = OpenAI(
    base_url="https://api.groq.com/openai/v1",  # swap to any provider
    api_key="YOUR_FREE_KEY",                     # freellms.org
)

response = client.chat.completions.create(
    model="llama-3.3-70b-versatile",
    messages=[{"role": "user", "content": "Hello via free AI proxy!"}],
)

Cursor / Claude Code / Codex CLI

Step-by-step proxy setup for every major tool: freellms.org

---

Why use a free AI proxy?

• Avoid vendor lock-in — swap providers by changing one URL
• Zero cost — all upstream providers have permanently free tiers
• OpenAI-compatible — works with Cursor, Codex, Claude Code, Aider, Cline, Open WebUI
• Fallback & failover — chain multiple free providers for resilience

Browse all proxy-ready providers, get free keys, and copy config snippets at freellms.org.

---

Related Resources

• Free LLM API keys
• Free OpenAI-compatible API
• Free LLM models
• Awesome free AI agents
• Awesome free LLM APIs

License

MIT — provider data sourced from freellms.org, updated daily.