Is ElevenLabs Down?

No — ElevenLabs is up

Reachable from all 8 checked regions

Average response time: 410ms

Last checked 5 hours ago · checks run every 6 hours

Official status page: https://status.elevenlabs.io

ElevenLabs uptime

100%

Last 7 days

100%

Last 30 days

100%

Last 90 days

398ms

Avg response, 30 days

Measured from multiple regions every 6 hours. Percentages count only checks that returned an availability answer — 62 days measured so far. A dash means that window does not yet hold enough measured days to publish a figure.

30-day history

62-day clean streak

Sep 14 Today

No downtime Partial Downtime Not measurable No data

Reachability by region

Each region runs its own request from a different part of the world. A service can be up for one continent and down for another, which is usually the first sign of a routing or CDN problem.

arn

300ms

DNS 48ms TCP 6ms TLS 23ms TTFB 274ms

cdg

301ms

DNS 48ms TCP 2ms TLS 11ms TTFB 174ms

dfw

279ms

DNS 235ms TCP 2ms TLS 6ms TTFB 274ms

ewr

147ms

DNS 55ms TCP 8ms TLS 13ms TTFB 121ms

fra

313ms

DNS 34ms TCP 1ms TLS 14ms TTFB 176ms

gru

696ms

DNS 207ms TCP 3ms TLS 56ms TTFB 431ms

iad

79ms

DNS 0ms TCP 2ms TLS 7ms TTFB 60ms

jnb

1169ms

DNS 367ms TCP 1ms TLS 172ms TTFB 798ms

What ElevenLabs does

ElevenLabs provides text-to-speech and voice cloning through an API and a web studio, used for audiobooks, dubbing, game dialogue and voice agents that answer calls. Its products are grouped as ElevenAPI for direct generation, ElevenAgents for conversational agents and ElevenCreative for the studio tools, so an outage can affect automated pipelines and live agents differently.

What an outage looks like

API calls to generate speech return errors or hang, so batch jobs and build pipelines stall rather than fail fast. Voice agents stop responding mid-conversation or never pick up at all. The web studio loads but generation requests spin without producing audio. Dubbing and long-form jobs queued earlier stop progressing, and newly cloned voices are unavailable to requests referencing them.

What to do about it

Check status.elevenlabs.io, which reports ElevenAPI, ElevenAgents and ElevenCreative separately, so an agent outage can be told apart from a generation one. Applications calling the API should back off and retry rather than hammering, since a retry storm turns a partial outage into a rate limit. Cache generated audio wherever the same text is used repeatedly.

Is it down for everyone, or just you?

If this page says ElevenLabs 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

ElevenLabs outage FAQ

Is the ElevenLabs API down or is my key rejected?

An authentication problem returns a 401 and does so instantly, while an outage more often shows as a timeout, a 5xx, or requests that hang. Check status.elevenlabs.io, which lists ElevenAPI separately from the studio and agents. The measurement above probes a public ElevenLabs API endpoint from several regions, so it reflects the API surface rather than the marketing site.

Do I get charged for failed generations?

ElevenLabs does not state a blanket policy on its status page, so treat this as something to confirm against your usage dashboard after an incident rather than assume either way. Compare the character or credit count against the audio you actually received. If the numbers do not match, that is the concrete thing to raise with support, with timestamps from the incident.

Will my voice agent recover on its own?

Agents holding a live connection generally do not resume a conversation that was cut, so the caller has to start again. New sessions should connect normally once ElevenAgents returns to operational. If your agent fronts a phone line, having a fallback message or a human handoff configured is what stops an outage becoming silence on the call.

Should my application retry when ElevenLabs is down?

Retry, but with exponential backoff and a cap. Speech generation is expensive and slow relative to a normal API call, so aggressive retries during a partial outage add load and can push you into rate limiting on top of the incident. Queue the work and drain it once the status page clears, rather than blocking a user request on a retry loop.

Does an outage affect voices I already cloned?

Cloned voices are stored on ElevenLabs rather than in your application, so they are not lost, but they cannot be used while the API is unavailable. Audio you have already generated and downloaded is unaffected. If a small set of phrases is critical to your product, generating and storing them ahead of time removes the dependency for that path.

How we measure this

We request ElevenLabs's public endpoint every 6 hours from Fly.io regions across six continents — the most recent check ran from 8 of them.
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. We only change the 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 ElevenLabs publishes an official status feed we read it too. An all-clear from the vendor can soften an unconfirmed degradation; a vendor-declared outage only worsens our verdict when our own checks corroborate it.

Get alerted when ElevenLabs goes down

This page refreshes every 6 hours. 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