Is Hugging Face down? Live Hub & Inference status
The open ML platform: Hub, Inference, Spaces
Live Hugging Face status — Hub, Serverless Inference API, and Spaces endpoints checked from your browser in real time.
huggingface.health provides real-time status and latency telemetry for Hugging Face services, including the model Hub, Serverless Inference API, Inference Endpoints, Spaces, and Datasets. It measures availability and response times directly from client browsers and edge probes.
Hugging Face API Status
Live, in-browser checks against Hugging Face's public APIs. Refreshes every 15s.
Hub
huggingface.co core API and marketing surface
- Hub Homepagehuggingface.co reachability—mswarming up
- Models APIPublic model listing endpoint—mswarming up
- Datasets APIPublic dataset listing endpoint—mswarming up
Serverless Inference
api-inference.huggingface.co edge
- Inference APIServerless inference gateway—mswarming up
- Routerrouter.huggingface.co OpenAI-compatible edge—mswarming up
Hugging Face doesn't return permissive CORS on every endpoint, so some checks confirm DNS/TCP/TLS reachability only. For component-level detail (Hub, Spaces, Inference Endpoints, Enterprise) see the official status page.
Hey, I'm Hugo — your Hugging Face guide. Ask about the Hub, Inference API, tokens, or rate limits.
Open Hugging FaceSponsored content. huggingface.health may earn a referral reward when you sign up via this widget. Answers are informational only — not professional advice.
Official Hugging Face status page
Open ↗About Hugging Face
Hugging Face is the largest open ML platform — a Hub of 1M+ models and datasets, a serverless Inference API, hosted Spaces, and enterprise Inference Endpoints. This page polls the public Hub and Inference edges every 15 seconds.
Hugging Face at a glance
- Founded
- 2016
- HQ
- New York, NY
- Models on Hub
- 1M+
- Datasets
- 250k+
- Spaces
- 500k+
- Inference options
- Serverless, Endpoints, Enterprise
Common Hugging Face issues
- Model loading (503 with estimated_time)
Serverless Inference lazy-loads models. Your first request after cold triggers a load; retry after the estimated time. Warm models respond in <1s.
- Rate-limit 429 on free tier
Anonymous and free-tier tokens hit stricter quotas. Add an HF_TOKEN with a Pro or Enterprise plan, or route through Inference Endpoints for dedicated capacity.
- Git LFS push fails on large uploads
Large model weights push through Git LFS. Timeouts usually mean network egress, not a Hub outage — resume with `huggingface-cli upload` for multipart resilience.
Background
Hugging Face started in 2016 as a chatbot company and pivoted into open-source NLP tooling, releasing the transformers library in 2019. It has since become the default hub for the open ML ecosystem — hosting model weights, datasets, demo Spaces, and both serverless and dedicated inference infrastructure used by hundreds of thousands of developers.
Frequently asked
How do I check if Hugging Face is currently down?
Why is Hugging Face Serverless Inference returning 503 errors?
Is huggingface.health affiliated with Hugging Face, Inc.?
How can I check if Hugging Face is down?
Why is the Hugging Face Inference API returning 503 or 504?
Are Hugging Face Spaces down or just sleeping?
About this check
All data is provided on a best-effort basis with no guarantees about accuracy or availability. This is an unofficial status page and is not affiliated with, endorsed by, or sponsored by Hugging Face, Inc.
Checks run client-side from your browser against Hugging Face's public endpoints. Latency reflects your network path, not the server's health alone. See our resources for more on how to read these checks.
Related on this site
All tracked surfaces
Live checks across the Hub and its sibling products.