Skip to main content
DeepInfra logo

Is DeepInfra Down?

No — DeepInfra is up
DeepInfra's status page
Not available · we could not read it
FlareWarden checks
Website reachable · 8 of 8 regions · 457 ms

Right now

We could not read DeepInfra's status page, so the verdict rests on our checks. Check it yourself How we decide

FlareWarden checks

api.deepinfra.com from 8 regions · when you open this page (at most every 30 minutes), otherwise every 6 hours

Reachable
Regions up
8 of 8
Response
457 ms
Checked

By region

  • Amsterdam reachable, 436 ms
  • Chicago reachable, 218 ms
  • Paris reachable, 423 ms
  • San Jose reachable, 93 ms
  • Singapore reachable, 1020 ms
  • Stockholm reachable, 518 ms
  • Sydney reachable, 704 ms
  • Toronto reachable, 249 ms
Timing by region
Connection timing for each region that loaded the page
RegionDNSTCPTLSTTFB
Amsterdam49 ms125 ms132 ms435 ms
Chicago117 ms30 ms35 ms218 ms
Paris43 ms124 ms129 ms423 ms
San Jose22 ms21 ms26 ms92 ms
Singapore415 ms198 ms204 ms1020 ms
Stockholm60 ms150 ms155 ms518 ms
Sydney208 ms159 ms165 ms704 ms
Toronto109 ms43 ms48 ms248 ms
We check that api.deepinfra.com loads, not every feature behind it. Test it yourself

History

DeepInfra website availability

Measured by FlareWarden · api.deepinfra.com

Last 7 days
100%
Last 30 days
100%
Last 90 days
100%
Avg response, 30 days
452ms

Last 30 days

21-day clean streak
Oct 7: no data
Oct 8: no data
Oct 9: no data
Oct 10: no data
Oct 11: no data
Oct 12: no data
Oct 13: no data
Oct 14: no data
Oct 15: no data
Oct 16: 100.00% uptime, 22 checks
Oct 17: 100.00% uptime, 29 checks
Oct 18: 100.00% uptime, 31 checks
Oct 19: 100.00% uptime, 32 checks
Oct 20: 100.00% uptime, 32 checks
Oct 21: 100.00% uptime, 32 checks
Oct 22: 100.00% uptime, 32 checks
Oct 23: 100.00% uptime, 32 checks
Oct 24: 100.00% uptime, 32 checks
Oct 25: 100.00% uptime, 32 checks
Oct 26: 100.00% uptime, 32 checks
Oct 27: 100.00% uptime, 32 checks
Oct 28: 100.00% uptime, 32 checks
Oct 29: 100.00% uptime, 32 checks
Oct 30: 100.00% uptime, 32 checks
Oct 31: 100.00% uptime, 32 checks
Nov 1: 100.00% uptime, 31 checks
Nov 2: 100.00% uptime, 32 checks
Nov 3: 100.00% uptime, 32 checks
Nov 4: 100.00% uptime, 32 checks
Nov 5: 100.00% uptime, 24 checks
Oct 7 Today
No downtime Partial Downtime Not measurable No data

Measured by FlareWarden's own checks of api.deepinfra.com from multiple regions every 6 hours. DeepInfra's status page sets the verdict above when it reports trouble, but never these figures. Percentages count only checks that returned an availability answer — 21 days measured so far. A dash means that window does not yet hold enough measured days to publish a figure.

What DeepInfra does

DeepInfra is an inference cloud that serves open source models, including LLMs, embeddings, rerankers, image and video generation, speech recognition and text to speech, billed per token. Developers point the OpenAI SDK at api.deepinfra.com/v1/openai and keep their existing code. It also runs private deployments of custom or fine-tuned models on dedicated GPUs and rents out B200 and B300 GPU clusters.

What an outage looks like

Most incidents hit one model. DeepInfra's status page tracks every model separately, and past downtime has kept a single model down for hours while the API row stayed at 100%. Calls to that model fail or time out while others answer. A 429 with a Rate limited message means you passed 200 concurrent requests on that model, though a very busy model can return 429 below it.

What to do about it

Open status.deepinfra.com and find the row for your exact model, not just the API and Website rows at the top. Retry a 429 after a short delay; DeepInfra says autoscaling adds capacity to a busy model shortly. If one model stays down, switch the model name to another in the catalog. The status page also lists scheduled deprecations. For help, ask in the DeepInfra Discord or email feedback@deepinfra.com.

Is it down for everyone, or just you?

If this page says DeepInfra is up but it is not loading for you, the problem is between you and them. Run a check against any URL from up to 17 cities across three regions to find out where it breaks.

Test it yourself

Related services

Together AI
Together AI
Fireworks AI
Fireworks AI
O
OpenRouter
Replicate
Replicate

DeepInfra outage FAQ

Is DeepInfra down, or just my model?
Usually just the model. DeepInfra's status page has rows for the API and the website, then a separate row with 90-day uptime for every model it monitors. Its history shows single models down for hours while the API row stayed at 100%, so a green banner does not clear your model. Search the model list on status.deepinfra.com for the exact name in your request.
What does a 429 from DeepInfra mean?
Usually that you went over 200 concurrent requests on one model, which is DeepInfra's default per-model limit. It counts requests in flight, not requests per minute, so slow responses reach it sooner. DeepInfra adds that a very busy model can return 429 even under the limit while autoscaling catches up. Retry after a short delay, and request a higher limit from Dashboard, Account.
What do the checks on this page measure?
They probe api.deepinfra.com, the API host, whose root answers with a short JSON greeting rather than a model response. A green check means the API host is reachable and answering from our regions; it does not prove that a given model is generating. For a model-level answer, use the row for that model on status.deepinfra.com, which tracks each model separately.
Can I switch to another provider during a DeepInfra outage?
Yes, if your code uses the OpenAI SDK. DeepInfra documents setting base_url to https://api.deepinfra.com/v1/openai with a DeepInfra token, so moving to another OpenAI-compatible provider means changing the base URL, the key and the model name. When only one model is affected, switching to a different model in the DeepInfra catalog is the quicker fallback and keeps your account and billing unchanged.
Why did a DeepInfra model name stop working?
It may have been deprecated rather than broken. DeepInfra's status page has a scheduled deprecations section that lists each retiring model, the date and time it goes, and the model that replaces it. Check there before treating the failure as an outage, and move your requests to the named replacement. If the model is not listed and still fails, check its own row on the status page.

How we measure this

  • We request DeepInfra's public endpoint every 6 hours from Fly.io regions across six continents — the most recent check ran from 8 of them. When you open this page and our last check is more than 30 minutes old, we check again on the spot; those checks update the verdict, never the availability figures.
  • A region counts as down only when it gets no usable HTTP response. A 403 or 429 means the origin answered and refused us, which we report as blocked, never as an outage.
  • A single failing region is treated as probe noise. Our checks only change their verdict when two consecutive cycles agree.
  • Response times average only the regions that actually served the page, so a timeout never inflates the number.
  • Where DeepInfra publishes an official status feed we read it too, and the verdict is the worse of what it reports and what our checks see: a major outage of DeepInfra itself is Down, trouble with one part of it is Degraded, problems at a few of a provider's many locations change nothing, and a security notice that marks nothing down is shown but changes nothing either. A reading counts for 45 minutes; we read the feed again within minutes of any change to its incidents. The availability figures above are our measurement alone.

Get alerted when DeepInfra goes down

We check DeepInfra every 6 hours, and again when you open this page if that was more than 30 minutes ago. Your own monitors run as often as every 30 seconds, from up to 17 cities across three regions, and tell you the moment something breaks.

Start Free Monitoring
Free plan available No credit card required