sevanalysis

Luma AI

AI. Source: Luma AI status page.

This status page publishes only incidents that are still open, so this page shows only incidents that were open when the collector checked, every three hours.

History starts 2026-09-02 17:00 UTC, the oldest incident this status page still shows. Figures for the last 90 days cover only the part since then.

Incidents, last 90 days
3+
Median to mitigation
not reported
Median to resolution
9h
p95 to resolution
13h
Naming an upstream provider
0%
With a postmortem on the status page
0%

What failed, last 90 days

The text does not say enough to tell2
Exhausted memory, disk, CPU, queues, connection pools or quotas1

Tagged by deepseek/deepseek-v4.1-flash from each incident's updates. See the methodology.

Incidents per month

Hatched months are before the recorded history starts.

All recorded incidents

Ray2 increased queue times minor 2026-09-16 12:00 UTC 13h

Affected: API, Dream Machine

Failure mode: Exhausted memory, disk, CPU, queues, connection pools or quotas. Root cause: Not enough capacity for the load. Tagged by deepseek/deepseek-v4.1-flash with medium confidence (taxonomy v0.3).

  1. degraded

    we are observing high queue times for Ray2

  2. resolved

    service is back to normal

Excerpts are shortened. Read the full incident on Luma AI's status page.

Ray2 Flash service degraded minor 2026-09-15 22:00 UTC 2h

Affected: API, Dream Machine

Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by deepseek/deepseek-v4.1-flash with medium confidence (taxonomy v0.3).

  1. degraded

    we’ve observed that ray2-flash jobs are timed out.

  2. resolved

    service is back to normal

Excerpts are shortened. Read the full incident on Luma AI's status page.

Ray2 Flash service degraded minor 2026-09-02 17:00 UTC 9h

Affected: API, Dream Machine

Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by deepseek/deepseek-v4.1-flash with high confidence (taxonomy v0.3).

  1. degraded

    we’ve observed that ray2-flash jobs are timed out.

  2. resolved

    Service is back to normal

Excerpts are shortened. Read the full incident on Luma AI's status page.