Hugging Face Inference Endpoints
Plans
| Plan | Billed and limits | Price |
|---|---|---|
| Self-Serve | Monthly; compute is priced by the minute, with listed instance rates shown per hour. · Pay as you use · Starting at $0.06/hour on product page · Email support | $0.06 a month |
| Enterprise | Custom annual contracts; contact Hugging Face for a quote. · Volume-based lower marginal costs · Uptime guarantees · Dedicated support and SLAs | On request |
As endpoints.huggingface.co lists them, read 2 Oct 2026. The maker's own page has the final word.
Facts from Hugging Face Inference Endpoints's own pages
- Access requirement
- Access to the Inference Endpoints web application requires a valid payment method on the Hugging Face account or organization. · huggingface.co, 2 Oct 2026
- API access
- Endpoints can be called through the UI, cURL, the Hugging Face inference libraries, or another REST client. · huggingface.co, 2 Oct 2026
- Autoscaling
- Users can configure minimum and maximum replicas and enable scale-to-zero when inactive. · huggingface.co, 2 Oct 2026
- Cloud providers
- The configuration guide lists AWS, Microsoft Azure, and Google Cloud Platform as hosting providers. · huggingface.co, 2 Oct 2026
- Compliance
- Hugging Face states that the Hub and Inference Endpoints are SOC 2 Type 2 certified. · huggingface.co, 2 Oct 2026
- Hub integration
- Endpoints use model weights and artifacts from Hugging Face Hub repositories. · huggingface.co, 2 Oct 2026
- Inference engines
- Native engine support includes vLLM, Text Generation Inference (TGI), SGLang, llama.cpp, and Text Embeddings Inference (TEI), with custom containers also supported. · huggingface.co, 2 Oct 2026
- Monitoring
- The web application provides endpoint logs and a metrics dashboard for monitoring deployments. · huggingface.co, 2 Oct 2026
- Notable limits
- Available instance types may require a quota request, and paused endpoints do not count against used quota while scale-to-zero endpoints do. · huggingface.co, 2 Oct 2026
- Private networking
- Private Endpoints are accessible only through an intra-region AWS or Azure PrivateLink connection and are not internet-accessible. · huggingface.co, 2 Oct 2026
- Purpose
- Inference Endpoints is a managed service for deploying AI models to production on infrastructure Hugging Face manages. · huggingface.co, 2 Oct 2026
- Security
- Hugging Face says endpoint payloads and tokens are not stored, logs are retained for 30 days, and traffic is encrypted in transit with TLS/SSL. · huggingface.co, 2 Oct 2026
- Support
- The product page lists email support for Self-Serve and dedicated support and SLAs for Enterprise. · endpoints.huggingface.co, 2 Oct 2026
Where Hugging Face Inference Endpoints stands on Freedom251
- AI Music APIs#12 of 21
- Model Hosting#15 of 25
- Llm Developer Tools#25 of 26