Free Hugging Face API Key, Base URL & Rate Limits
Model ProviderHugging Face provides free access to a rotating selection of open-weight models including Qwen, Llama, Gemma, and SmolLM.
How to get a free Hugging Face API key
- 1
- 2 Go to Settings → Access Tokens
- 3 Create a token (read-only is fine)
- 4 Pick a model Free models are rate-limited on shared infrastructure.
- 5 Configure OpenAI client Base URL: https://router.huggingface.co/v1
Provider Snapshot
Supported Models 7 models
View in directory →| Model ID | Developer | Context | Availability | Free Tier | Use cases |
|---|---|---|---|---|---|
| qwen2-5-7b-instruct Qwen2.5-7B-Instruct | Hugging Face | 131K | Online | Yes | chat |
| qwen2-5-coder-7b-instruct Qwen2.5-Coder-7B-Instruct | Alibaba | 131K | Online | Yes | chatcoding |
| meta-llama-3-1-8b-instruct Meta-Llama-3.1-8B-Instruct | Meta | 128K | Online | Yes | chat |
| gemma-3-4b-it gemma-3-4b-it | 131K | Online | Yes | chat | |
| mixtral-8x7b-instruct-v0-1 Mixtral-8x7B-Instruct-v0.1 | Hugging Face | 32K | Check provider | Yes | chat |
| phi-3-5-mini-instruct Phi-3.5-mini-instruct | Hugging Face | 128K | Check provider | Yes | chat |
| mistral-7b-instruct-v0-3 Mistral-7B-Instruct-v0.3 | Hugging Face | 32K | Check provider | Yes | chat |
Developer Tools
Related Providers
View all →Hugging Face FreeLLM Score free API access score
How we score →What is Hugging Face?
Hugging Face Inference API — Qwen, Llama, Gemma at ~1,000 RPD.
Hugging Face provides free access to a rotating selection of open-weight models including Qwen, Llama, Gemma, and SmolLM. The free tier is rate-limited (~1,000 requests/day) and uses shared infrastructure, so latency varies. The live-tested router endpoint supports OpenAI-compatible Chat Completions and Responses.
- Rotating selection of open models
- ~1,000 RPD free tier
- No credit card required
- OpenAI-compatible router endpoint
API Compatibility: Live-tested OpenAI-compatible Chat Completions and Responses
Hugging Face Free Tier Limits & Pricing
Hugging Face API Setup Tutorials
Hugging Face 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 →
Hugging Face Model Use Cases
What Hugging Face's free models are best for, based on aggregated model capabilities:
Hugging Face Limitations & Caveats
- Cold starts common — first request may take 30s+
- Models larger than 10GB may fail to load on free tier
- No SLA — shared infrastructure, availability not guaranteed
Hugging Face FAQ
Why is my first Hugging Face API request so slow?
Free tier uses serverless inference with cold starts. The first request to a model loads it from disk, taking 30-60 seconds. Subsequent requests within ~15 minutes are fast. Use a keepalive ping to avoid cold starts.
Can I use the Hugging Face Inference API with OpenAI SDK?
Yes. The live-tested router endpoint at https://router.huggingface.co/v1 supports OpenAI-compatible Chat Completions and Responses for supported models. Some older Hugging Face inference endpoints still use Hugging Face-specific request formats.
Which models are actually free on Hugging Face?
Any model with the "Inference API" tag is available for free on the serverless tier. However, rate limits apply (~1,000 RPD) and larger models (>10GB) may not load. Popular free models include Qwen, Llama, Gemma, and SmolLM.