AI APIs / LLM API / Cohere

Cohere

OPERATIONAL

Since Mar 2025 · 25 incidents · status.cohere.com · checked 3m ago

CompareStatus page
GradeCLast 90 days
CPRODA+SHIP

Overview

30d90d1y
Admitted uptimeTime not covered by a Sev 1 the provider posted itself. A best case.
99.985%Grade C
IncidentsEverything posted on the status page except scheduled maintenance.
81 Sev 1
Sev 1 downtimeTotal minutes of Sev 1 incidents, as posted.
1h 18m
Median fixMedian time to resolve. Half of incidents were fixed faster.
1h 18m
LongestLongest single incident, start to resolved.
4h 22m
Since lastTime since the last counted incident was resolved.
51d 0h

By area

CProduction
4INCIDENTS
9h 59mAFFECTED
Your users are hit: apps, APIs, databases, inference
Worst, 1y
SEV 1Feb 13, 2026 · 1h 18mSome chat models are unresponsive or timing out
SEV 2Sep 1, 2026 · 4h 22mMultiple Endpoint Disruptions
SEV 2Jun 1, 2026 · 2h 54m/audio/transcriptions endpoint returns error for audio files with unexpected sampling rate
2 in the last 6 months, 2 in the 6 before.
A+Shipping
1INCIDENTS
51mAFFECTED
You can't deploy, build or manage: CI, CLI, dashboard, auth, billing
Worst, 1y
SEV 3Jul 23, 2026 · 51mDashboard Login Service Disruption
Worse. 1 in the last 6 months, 0 in the 6 before.
CleanSev 3–4Sev 1–2No data

Trend, last 12 months

4 in the last 6 months, 4 in the 6 before.

Severity mix

Sev 1 · Lights outProd site, API or database down, or data loss1
Sev 2 · BleedingProd degraded or partly down3
Sev 3 · Can't shipCan't deploy, build or manage it1
Sev 4 · PapercutDocs, status page, slow dashboard3

When it breaks

Start times, last 12 months, UTC.

What breaks

No components named in the last 12 months.

Communication

Updates eachStatus page updates per incident, on average.
3.4Last 12 months
Update gapMedian time between status page updates during an incident.
16mLast 12 months
PostmortemsSev 1–2 incidents with a published write-up.
0 of 4Sev 1–2

Incident history

Sep 1, 20264h 22mSEV 2
Multiple Endpoint DisruptionsCohere endpoints were disrupted from about 15:41 to 20:03 UTC on September 1, 2026. Cohere traced the root cause to an ongoing Google Cloud Platform outage and monitored until GCP resolved it. The update text does not specify which endpoints or how many requests failed.AI-graded
Why this grade
  1. RULES · SEV 2
    Sev 2 from the incident title (no components listed)
  2. CLAUDE · SEV 2 · UNCLEAR · LOW CONFIDENCE
    Multiple Cohere endpoints were disrupted for about four and a half hours due to a Google Cloud outage; extent across endpoints is not stated.
  3. Claude Haiku 5.5 · prompt v5 · Oct 22, 2026 · Raw record
4 updates
  1. 15:41 UTC · investigating
    🔎 We are currently investigating this live incident.
  2. 16:21 UTC · identified
    🛠️ We have identified the root cause as an ongoing Google Cloud Platform (GCP) outage. Our team is actively monitoring the situation while we wait for GCP to resolve the underlying issue. ([reference](https://status.cloud.google.com/incidents/J5ia5t9p3g9Q5Wi7r8Ev "reference"))
  3. 19:47 UTC · monitoring
    👀 We are monitoring to make sure the incident has been fully resolved.
  4. 20:03 UTC · resolved
    🎉 The issue has been resolved and we are back to normal.
Source
Aug 5, 202657mSEV 4
Documentation Content Loading IssuesSome Cohere documentation pages failed to load for about an hour due to an upstream Fern issue; the API was not described as affected.AI-graded
Why this grade
  1. RULES · SEV 4
    Sev 4 from the incident title (no components listed)
  2. CLAUDE · SEV 4 · UNCLEAR · MEDIUM CONFIDENCE
    Some Cohere documentation pages failed to load for about an hour due to an upstream Fern issue; the API was not described as affected.
  3. Claude Haiku 5.5 · prompt v5 · Oct 22, 2026 · Raw record
3 updates
  1. 19:36 UTC · investigating
    We are investigating a partial outage affecting our documentation site. Some documentation pages may not load correctly due to an ongoing issue with our upstream documentation provider (Fern). We are monitoring the incident and will share updates as they become available.
  2. 20:33 UTC · resolved
    🎉 The issue has been resolved and we are back to normal.
  3. 20:35 UTC · resolved
    🎉 The issue has been resolved and we are back to normal.
Source
Jul 23, 202651mSEV 3
Dashboard Login Service DisruptionCohere dashboard login showed a continuous refresh loop, preventing successful authentication for about 50 minutes. The cause was identified and fixed, and the provider said API access and model inference were unaffected throughout.AI-graded
Why this grade
  1. RULES · SEV 3
    Sev 3 from the incident title (no components listed)
  2. CLAUDE · SEV 3 · UNCLEAR · HIGH CONFIDENCE
    Dashboard login looped and blocked authentication for about 50 minutes; the provider says API and model inference were unaffected.
  3. Claude Haiku 5.5 · prompt v5 · Oct 22, 2026 · Raw record
4 updates
  1. 22:56 UTC · investigating
    🔎 We are currently investigating an issue affecting the Cohere Dashboard ([dashboard.cohere.com](http://dashboard.cohere.com "dashboard.cohere.com")). Users may experience a continuous refresh loop on the login page, preventing successful authentication.
  2. 23:27 UTC · identified
    🛠️ **Update:** We have identified the cause of login issues affecting the Cohere Dashboard, preventing successful logins. A fix is being implemented now and we expect Dashboard login to be restored shortly. **Note:** API access, model inference, and all other Cohere services are **unaffected** and operating normally.
  3. 23:36 UTC · monitoring
    👀 The issue affecting Dashboard login has been resolved ✅. We're monitoring to confirm stability. All Cohere services are operating normally.
  4. 23:46 UTC · resolved
    🎉 The issue affecting Dashboard login has been resolved. All Cohere services are operating normally. Thank you for your patience while we worked through this.
Source
Jun 1, 20262h 54mSEV 2
/audio/transcriptions endpoint returns error for audio files with unexpected sampling rateThe audio transcription endpoint rejected audio at the expected sampling rate for about three hours; other endpoints were not reported affected.AI-graded · some customers
Why this grade
  1. RULES · SEV 2
    Sev 2 from the incident title (no components listed)
  2. CLAUDE · SEV 2 · SUBSET · MEDIUM CONFIDENCE
    The audio transcription endpoint rejected audio at the expected sampling rate for about three hours; other endpoints were not reported affected.
  3. Claude Haiku 5.5 · prompt v5 · Oct 22, 2026 · Raw record
2 updates
  1. 13:51 UTC · identified
    /audio/transcriptions endpoints availability is affected now due to the Cohere ASR model backing the endpoint throws error for any audio inputs that are in the expected sampling rate. Previously these sampling rate audio files were accepted and processed, marking this as a regression type incident. :hammer_and_wrench: We have identified the root cause of the incident, and are working diligently to fix.
  2. 16:46 UTC · resolved
    🎉 The issue has been resolved and we are back to normal.
Source
Apr 20, 202613mSEV 4
Docs Platform Degraded PerformanceCohere's documentation site was down for about 13 minutes due to a third-party docs provider issue; the API was not reported affected.AI-graded
Why this grade
  1. RULES · SEV 4
    Sev 4 from the incident title (no components listed)
  2. CLAUDE · SEV 4 · UNCLEAR · HIGH CONFIDENCE
    Cohere's documentation site was down for about 13 minutes due to a third-party docs provider issue; the API was not reported affected.
  3. Claude Haiku 5.5 · prompt v5 · Oct 22, 2026 · Raw record
2 updates
  1. 22:16 UTC · investigating
    We are currently experiencing an outage affecting our documentation site due to an issue with our third-party documentation provider, Fern. Our team is actively monitoring the situation while the provider investigates and works on a fix.
  2. 22:28 UTC · resolved
    Docs are back up!
Source
Apr 13, 20261h 25mSEV 2
Multiple Model DisruptionsSev 2 from the incident title (no components listed)
3 updates
  1. 19:36 UTC · investigating
    We are currently investigating Multiple Model Disruptions.
  2. 19:45 UTC · monitoring
    👀 We are monitoring to make sure the incident has been fully resolved.
  3. 21:00 UTC · resolved
    🎉 The issue has been resolved and we are back to normal.
Source
Apr 11, 20269mSEV 4
Scheduled DB Maintenance: Model authentication and API access momentarily affectedSev 4 from the incident title (no components listed)
2 updates
  1. 15:31 UTC · identified
    There is a scheduled DB maintenance in progress. All the API authentication when accessing models are affected momentarily. The auth and API services is coming back online and should be fully up very soon.
  2. 15:39 UTC · resolved
    🎉 The issue has been resolved and we are back to normal.
Source
Feb 13, 20261h 18mSEV 1
Some chat models are unresponsive or timing outSev 1 from the incident title (no components listed)
3 updates
  1. 04:34 UTC · investigating
    We are currently investigating an incident with chat models.
  2. 05:35 UTC · monitoring
    We identified and resolved the connectivity problem affecting certain chat models. Affected models are starting to come back online.
  3. 05:51 UTC · resolved
    The issue has been resolved and all models are back to normal operation
Source
Oct 18, 202537mSEV 2
Chat models return 503 service error when /v1/chat requests omit message and chat_history while providing tool_resultsSev 2 from the incident title (no components listed)
2 updates
  1. 21:46 UTC · identified
    🛠️ We have identified the root cause of the incident, and are working diligently to fix.
  2. 22:23 UTC · resolved
    🎉 The issue has been resolved and we are back to normal.
Source
Oct 1, 20253h 50mSEV 2
logprobs param degeraded for `command-a-03-2025` and `command-a-reasoning-08-2025`Sev 2 from the incident title (no components listed)
2 updates
  1. 15:56 UTC · investigating
    We have identified that the `logprobs` parameter is not working as expected for the following models: • `command-a-03-2025` • `command-a-reasoning-08-2025` The team is investigating the incident.
  2. 19:46 UTC · resolved
    `logprobs` feature has been correctly marked as unavailable for the specified models.
Source

Peers, 90 days

Compare
GroqGrade B100.00% up3 incidentsMTTR 23mCohereGrade C100.00% up2 incidentsMTTR 4h 22mOpenAI APIGrade C100.00% up27 incidentsMTTR 1h 38mAnthropic APIGrade D99.97% up50 incidentsMTTR 1h 13m

README badge

Cohere's live grade, for your README.

Cohere grade badge
[![cohere on downagain.dev](https://downagain.dev/badge/cohere.svg)](https://downagain.dev/p/cohere)