Stack Health Score
Score = 30-day uptime − penaltiesLatency >500ms: −2 | >1000ms: −5
Outages this month: −2 each
Degradations this month: −1 each
Uptime · Latency · Incidents
Does your product depend on Hugging Face Inference?
Get alerted the moment Hugging Face Inference degrades — before your users notice. Monitor it alongside your whole stack, free for 30 days.
99.90%
UptimePercentage of time the API was available.
Only DOWN status impacts this metric.
155ms
Avg Latency
1,000+
APIs monitored
1
Outages
0
Degradations
This Month
About Hugging Face Inference's Hugging Face Inference
Hugging Face Inference's Hugging Face Inference runs thousands of open-source AI models via a unified cloud inference API. Hugging Face was founded in 2016 and hosts the world's largest open-source AI model hub.. The Hugging Face Inference is used by developers worldwide for model inference, text generation, image classification, embeddings, NLP tasks.
Probecast monitors the Hugging Face Inference from 4 global regions (US West, US East, Europe, Asia Pacific) every 300 minutes, tracking uptime, latency and regional availability. Our real-time monitoring helps developers and teams understand the current health of Hugging Face Inference's Hugging Face Inference and get alerted the moment service degrades.
Why monitor Hugging Face Inference?
If your product depends on Hugging Face Inference for critical features, an outage on Hugging Face Inference's side is an outage on your product. Probecast detects Hugging Face Inference incidents before your users do, giving you time to communicate with customers, fall back to alternative services or pause background jobs gracefully.
Current status
As of the latest check, Hugging Face Inference is operational. The 30-day uptime is 99.90%, with 1 incident recorded this month. Average response latency across all regions is 155ms.
●Uptime History
Regional Status
US West
103ms
latency
US East
98ms
latency
Europe
121ms
latency
Asia Pacific
299ms
latency
Latency (24h)
Hugging Face Inference is one of 1,000+ services Probecast watches for you
Checking a status page during an incident means your users already noticed. Probecast flips that: it tells you first — and shows what the outage costs.
Checked from 4 regions
US West & East, Europe and Asia — a regional outage never hides.
Alerted the moment it degrades
Slack, email, Teams or webhooks — before your users notice.
Your whole stack, one screen
1,000+ catalog APIs plus your own endpoints, websites and cron jobs.
Your business view: traffic & revenue
Google Analytics, Probecast Pixel, Stripe, Polar, Lemon Squeezy & more — beside your uptime.
2-minute setup · no DevOps team needed · cancel anytime during the trial
Frequently Asked Questions
Recent Incidents
API Information
Monitored Endpoint
https://huggingface.co/api/models?limit=1Provider
Category
Artificial Intelligence
Health Check
GET request
Check Interval
Every 5 minutes