How to Get a Free GitHub Models API Key (2026)
10 free models available — no credit card required. Get your GitHub Models API key → Test free models →
GitHub Models FreeLLM Score
A solid choice for most developers with balanced limits and model quality.
What is GitHub Models?
GPT-4o, o3, Llama 4, DeepSeek-R1 — free for all GitHub users.
GitHub Models provides free API access to 45+ models from OpenAI (GPT-4.1, o3, o4-mini), Meta (Llama 4), Mistral, DeepSeek, and Cohere for GitHub account holders. Rate limits depend on the GitHub Copilot subscription tier (Free/Pro/Pro+/Business). Tokens per request are limited (8K in/4K out), making it best suited for prototyping rather than production workloads.
- 45+ models including GPT-4.1 and o3
- Free for all GitHub accounts
- Includes Llama 4, DeepSeek-R1, Mistral
- Base URL: models.inference.ai.azure.com
API Compatibility: OpenAI SDK-compatible (Chat Completions)
How to Get a GitHub Models API Key
- 1 Sign in with GitHub account Every GitHub user gets free access.
- 2 Go to github.com/marketplace/models
- 3 Generate a personal access token with Models:read permission
- 4 Pick a model GPT-4.1, o3, Llama 4, DeepSeek — 45+ models available.
- 5 Configure OpenAI client Base URL: https://models.inference.ai.azure.com
All Free GitHub Models Models — Context Windows & Rate Limits
| Model | Context | Max Output | Modality | Rate Limit | Released | Status |
|---|---|---|---|---|---|---|
| o4-mini | 200K | 100K | 10 RPM, 50 RPD | Apr 16, 2025 | Online | |
| gpt-4.1 | 1.0M | 32K | 10 RPM, 50 RPD | Apr 14, 2025 | Online | |
| gpt-4.1-mini | 1.0M | 32K | 15 RPM, 150 RPD | Apr 14, 2025 | Online | |
| gpt-5 | 200K | 32K | 10 RPM, 50 RPD | Aug 7, 2025 | Online | |
| Llama-4-Scout-17B-16E | 512K | 4K | 15 RPM, 150 RPD | — | Online | |
| Llama-4-Maverick-17B-128E | 256K | 4K | 10 RPM, 50 RPD | — | Online | |
| gpt-4o | 128K | 16K | 10 RPM, 50 RPD | May 13, 2024 | Online | |
| Mistral-Small-3.1 | 128K | 4K | 15 RPM, 150 RPD | Mar 17, 2025 | Online | |
| DeepSeek-R1 | 64K | 8K | 15 RPM, 150 RPD | May 28, 2025 | Online | |
| Meta-Llama-3.3-70B | 131K | 4K | 15 RPM, 150 RPD | Dec 6, 2024 | Online |
GitHub Models Free Tier Limits & Pricing
GitHub Models API Setup Tutorial & Tools
GitHub Models is fully compatible with popular AI coding assistants like Cursor, Claude Code, and more. To see step-by-step API configuration instructions for your favorite tool, please visit our Global Configuration Guide →
Use Cases
What GitHub Models's free models are best for, based on aggregated model capabilities:
Limitations & Caveats
- Low per-request token limits (8K input / 4K output)
- Rate limits tied to GitHub Copilot subscription tier
- Not suitable for large-context or long-generation tasks
Frequently Asked Questions
How many requests can I make with GitHub Models free tier?
Rate limits depend on your GitHub Copilot subscription: Free tier gets ~10 requests/minute, Pro gets ~20 RPM, and Pro+/Business get higher limits. The 8K input / 4K output token limit applies to all tiers.
Can I use GPT-4.1 or o3 on GitHub Models for free?
Yes — GitHub Models is one of the few places offering free access to OpenAI's latest models including GPT-4.1, o3, and o4-mini. However, the low token limits (8K in/4K out) make it best for prototyping.
Why is my GitHub Models request getting rate limited?
Rate limits are tied to your Copilot subscription tier. If you're on the free tier, you get ~10 RPM. Upgrade to Copilot Pro for ~20 RPM, or switch to another provider for higher limits.