Hugging Face Inference API

The Hugging Face Inference API allows you to run inference on 100,000+ open-source models for free.

🚀 Key Specs

  • Model Support: Llama 3, Mistral, Gemma, Stable Diffusion, Whisper, and many more.
  • Rate Limit: Shared rate limits per user token. If a model is high traffic, you may experience downtime.
  • Cost: Free to use with a Hugging Face account.
  • Best For: Testing open source LLMs and multi-modal models.

🗝 How to get an API Key

  1. Visit huggingface.co.
  2. Click on your Profile -> Settings -> Access Tokens.
  3. Create a “Read” token.

🛠 Usage Example

Check out hf-text-generation.py for the Python SDK version.