Free APILLMAPI keyOfficial

GitHub Models

Frontier and open models for free using only your GitHub token

GitHub Models gives every GitHub account rate-limited free inference against a catalogue of models (OpenAI GPT family, Llama, Mistral, Phi, and more) through an OpenAI-compatible endpoint. Auth is a GitHub personal access token with the models:read scope; no separate billing or card is needed for free usage.

View docs Get an API key

Free tier

Free for experimentation, tied to your GitHub account (no card). Opt in to paid usage for production-grade limits and larger context.

Rate limits

Free, per-account, varies by model tier (verified June 2026, Copilot Free): high-tier models 10 RPM / 50 req/day; low-tier models 15 RPM / 150 req/day; embeddings 15 RPM / 150 RPD. Per-request token caps apply (e.g. 8000 in / 4000 out on low models). Limits subject to change.

Auth

API key

Base URL

https://models.github.ai/inference

What you'd build with it

Prototype with GPT-class and open models using the GitHub token you already have
CI/automation steps in GitHub Actions that call a model with the built-in token
Cheap embeddings for a quick RAG proof-of-concept
A free fallback provider for a multi-model Claude Skill

Key endpoints

POST/chat/completionsOpenAI-compatible chat completion across catalogue models
POST/embeddingsGenerate text embeddings
GET/catalog/modelsList the available models in the catalogue

Members

Members get the runnable recipe

Create a free account Sign in