Skip to main content
← All Comparisons

Mistral vs OpenAI.

Live

Over the last 30 days CheckUpstream measured Mistral higher: 100.000% availability across 2307 checks, against OpenAI's 99.044% across 2720. Over the last 30 days Mistral posted 39 incidents and OpenAI posted 2.

Mistral
Degraded

30d measured availability

100.000%

Incidents posted (30d)

39

Days Mistral declared impact, last 30 days2026-08-02 · 100.00%2026-08-03 · 100.00%2026-08-04 · 100.00%2026-08-05 · 99.86%2026-08-06 · 99.99%2026-08-07 · 79.73%2026-08-08 · 70.00%2026-08-09 · 69.52%2026-08-10 · 70.00%2026-08-11 · 70.00%2026-08-12 · 87.39%2026-08-13 · 99.59%2026-08-14 · 98.11%2026-08-15 · 95.05%2026-08-16 · 100.00%2026-08-17 · 100.00%2026-08-18 · 100.00%2026-08-19 · 96.40%2026-08-20 · 100.00%2026-08-21 · 100.00%2026-08-22 · 100.00%2026-08-23 · 100.00%2026-08-24 · 100.00%2026-08-25 · 78.76%2026-08-26 · 42.42%2026-08-27 · 30.00%2026-08-28 · 53.01%2026-08-29 · 70.00%2026-08-30 · 70.00%2026-08-31 · 92.66%
30 days agotoday
OpenAI
Operational

30d measured availability

99.044%

Incidents posted (30d)

2

Days OpenAI declared impact, last 30 days2026-08-02 · 100.00%2026-08-03 · 100.00%2026-08-04 · 99.68%2026-08-05 · 100.00%2026-08-06 · 100.00%2026-08-07 · 100.00%2026-08-08 · 100.00%2026-08-09 · 100.00%2026-08-10 · 100.00%2026-08-11 · 100.00%2026-08-12 · 100.00%2026-08-13 · 100.00%2026-08-14 · 100.00%2026-08-15 · 100.00%2026-08-16 · 100.00%2026-08-17 · 100.00%2026-08-18 · 100.00%2026-08-19 · 100.00%2026-08-20 · 100.00%2026-08-21 · 99.86%2026-08-22 · 100.00%2026-08-23 · 100.00%2026-08-24 · 100.00%2026-08-25 · 100.00%2026-08-26 · 100.00%2026-08-27 · 100.00%2026-08-28 · 100.00%2026-08-29 · 100.00%2026-08-30 · 100.00%2026-08-31 · 100.00%
30 days agotoday

Disclosure by window

Share of each window with no impact declared, for Mistral and OpenAI, across five time windows
WindowMistralOpenAI
7d62.408%96.816%
30d85.750%97.420%
90d95.233%92.872%
180d97.616%95.419%
1y98.824%97.741%

Share of each window in which the vendor declared no customer impact, from their own status posts. A higher figure means fewer or shorter declared incidents — it is not a measurement of availability, and it does not make one vendor more reliable than the other. The measured availability figures are in the cards above, and only where we run a probe.

Mistral: recent incidents (30d)

Full scorecard →
Show 34 older incidents

OpenAI: recent incidents (30d)

Full scorecard →
  • Unexpected logouts for some ChatGPT web users

    Minor

    Aug 21, 19:48 UTC · resolved in 6m · Resolved

  • Elevated ChatGPT conversation errors affecting Plus, Pro, Business, and Edu users

    Major

    Aug 4, 12:43 UTC · resolved in 6m · Resolved