Is Hugging Face down?
Models, datasets and inference.
Severity over 90 days
From the incidents published by the provider. Maintenance excluded.
Downtime over 90 days
What to do if Hugging Face is down
Check it isn't your side
Open another site. If it loads, your connection is fine.
Read the official source
Hugging Face's status page is authoritative. It is the one we poll every 5 minutes.
Official status page →See what people are reporting
Outages often surface on X before Hugging Face publishes them.
“Hugging Face down” reports on X →Know when it is back
Rather than reloading this page, leave your address: the alert goes out as soon as it updates.
Incident history
Published by Hugging Face
Maintenance: Scheduled maintenance: Inference Endpoints (H200)
Maintenance completed
Maintenance: Scheduled maintenance: Inference Endpoints (H200)
We're performing infrastructure maintenance on H200 Inference Endpoints today until around 12:30 UTC. Creating new H200 endpoints may be temporarily limited.
Slower downloads in the Asia-Pacific region
The upstream network capacity issue affecting our Singapore CDN location has been resolved, and download performance for Asia-Pacific users is back to normal. We've returned traffic routing to its usual configuration. Thank you for your patience.
Slower downloads in the Asia-Pacific region
The Asia Pacific region continues to experience degraded performance due to impaired underlying network capacity. The root cause is identified as the rupture of a subsea cable between China and Japan, rerouting traffic onto the remaining capacity. We continue to monitor the situation and are diverting part of traffic to healthy regions.
Some GPU Jobs and Spaces failing to start
The issue is resolved. If your job failed between 18:00 UTC on September 24 and 09:10 UTC on September 25, please resubmit it. If your Space is still showing a Runtime error, restart it from the Space settings.
Some GPU Jobs and Spaces failing to start
Between 18:00 UTC on September 24 and 09:10 UTC on September 25, some GPU Jobs and GPU Spaces failed to start. Affected jobs ended with an ERROR status before the container started.
Slower downloads in the Asia-Pacific region
We're seeing slower download speeds for some users in the Asia-Pacific region who are served by our Singapore CDN location. Reduced network capacity between cloud regions is impacting our ability to download files that aren't already cached in the region. We are not seeing an increased error rate, so downloads should complete successfully. To reduce the impact, we're sending some Asia-Pacific tra
Maintenance: Scheduled maintenance: Hub
Maintenance completed
Maintenance: Scheduled maintenance: Hub
During a short window inside this period (roughly 20 minutes) write operations on the Hub may fail while we migrate database infrastructure. This includes creating repositories, pushing commits and uploading files, opening or commenting on pull requests and discussions, and changing account, organization or billing settings. It affects both the UI and the API. Reads are not affected: browsing the
Scheduled maintenance: Hub
Maintenance completed
Maintenance: Scheduled maintenance: Hub
During a brief window (approximately 5–10 minutes), uploads and commits to repositories may temporarily fail or need to be retried while we upgrade database infrastructure. This affects both the UI and the API. Downloads, reads, and browsing are not affected, and no action is needed on your end.
Maintenance: Scheduled maintenance: Spaces
During a brief window (approximately 5–10 minutes), write and management operations on Spaces (creating or duplicating a Space, changing hardware or settings, pausing/restarting, and updating secrets and variables) may temporarily return a 503 error while we upgrade infrastructure. This affects both the UI and the API. Reads and already-running Spaces are not affected, and no action is needed on y
Elevated error rate - AWS CDN (Singapore)
Our fix is now in place and this incident is resolved. Applying it caused one final short period of failed downloads between 23:00 and 23:14 UTC, as we noted could happen. Downloads from the Singapore region have been stable since, and we have added capacity to prevent this from recurring. We are sorry for the disruption.
Elevated error rate - AWS CDN (Singapore)
Some file downloads from our Singapore region (ap-southeast-1) failed with 500 errors during two short periods, starting at 15:14 UTC. Downloads are normal again now. The cause is not fully resolved, so short bursts of errors can return until our fix is in place. Other CDN regions are not affected. We are sorry for the disruption.
Maintenance: Scheduled maintenance: Inference Endpoints
During a short window within this period (roughly 5–10 minutes), endpoint management operations (creating, updating, pausing, or resuming endpoints) may briefly return a 503 ("temporarily unavailable due to maintenance") while we complete infrastructure upgrades. Running endpoints and inference requests are not affected, and no action is needed on your end.
Maintenance: Scheduled maintenance: Inference Endpoints
During a short window inside this period (roughly 5–10 minutes) endpoint management operations (creating, updating, pausing, or resuming endpoints) may briefly return a 503 ("temporarily unavailable due to maintenance") while we complete upgrades to infrastructure. Running endpoints and inference requests are not affected, and no action is needed on your end.
Hub unavailable
Services are back to nominal operations. Root cause confirmed to global Amazon Web Services outage.
Hub unavailable
We are mitigating the issue, and services are coming back up.
Hub unavailable
Hugging Face platform is unavailable from most regions in the world due following a global outage from Amazon Web Services. We are working on a mitigation.
Maintenance: Scheduled maintenance: Jobs
During a short window inside this period (roughly 5–10 minutes) write operations may briefly return a 503 ("temporarily unavailable due to maintenance") while we complete upgrades to infrastructure. Reads and already-running jobs are not affected, and no action is needed on your end.
Elevated error rate – AWS CDN (Singapore)
Resolved. Error rates returned to normal at 08:05 UTC and the CDN is operating normally. The disruption lasted approximately 7 minutes. We apologize for any inconvenience.
Elevated error rate – AWS CDN (Singapore)
Between 07:58 and 08:05 UTC, the AWS CDN returned an elevated rate of HTTP 500 errors for a portion of requests in the Asia-Pacific (Singapore / ap-southeast-1) region. The cause was a transient internal service disruption, which has been mitigated.
Data serving slowness in Asia region
Issue is solved, Asia traffic back to nominal bandwidth performances.
Data serving slowness in Asia region
We are experiencing slowness on large file downloads in Asia region since 7AM UTC. This only affects downloads via the huggingface CLI. Root cause identified and mitigation in progress.
Official source
Data from Hugging Face's public status page.
Official status pageBadge for your site
Show Hugging Face's status on your own site. Updates automatically, same address as this page.
<a href="https://www.upstreamstatus.com/en/status/hugging-face"><img src="https://www.upstreamstatus.com/badge/hugging-face?locale=en" width="280" height="64" alt="Hugging Face status" /></a>
Other artificial intelligence services
Frequently asked questions
Is Hugging Face down right now?
As of the last check (2026-11-01 00:45 UTC), Hugging Face reports “Operational” on its official status page. No ongoing incident is published.
What is Hugging Face's uptime over the last 90 days?
Hugging Face's history is still being built up on Upstream Status.
How do I get notified automatically when Hugging Face goes down?
Create a free alert on Upstream Status: as soon as an incident is published on Hugging Face's status page, you get an email (or a Slack/webhook notification on a paid plan).