Scaleway Generative APIs outage history

Biggest Scaleway Generative APIs outages

Ranked by the official impact level, then by how long they lasted. Each has its own page with the cause, the fix time and every official update.

1. Oct 21, 2026: [GAPI] - [global] - Dedicated deployment deletion failing 13 hours 24 min · rated "Major"

Every smaller Scaleway Generative APIs incident

Minor and short incidents from the Scaleway status page, newest first, each with a line from the official updates. Times are UTC.

October 2026

• Oct 21, 2026, 09:20 UTC · [Generative API] - [global] - New dedicated deployment creation temporarily unavailable

  7 hours 41 min · rated "Minor"

  "The issue has been identified and a fix is being implemented."

• Oct 20, 2026, 15:31 UTC · [Generative API] - [global] - GLM-5.3 timeouts

  ongoing, 45 hours 4 min so far · rated "Minor" · still open at our last refresh

  "Started Tuesday 20th October @ 5:04pm The glm-5.3 model is consistently returning timeout errors on all requests. We are working at fixing the issue."

• Oct 20, 2026, 13:59 UTC · [Generative API] - [global] - Bge-multilingual-gemma-2 became unavailable

  24 minutes · rated "Minor"

  "Started Tuesday 20th October 2:50pm The model is not available. We're working to fix it asap."

• Oct 19, 2026, 17:02 UTC · [Generative API] - [global] - Deepseek v4.1 flash timeouts

  20 hours 1 min · rated "Minor"

  "19th October from 2:30pm to 7pm Some requests to the deepseek-v4.1-flash model have experienced elevated timeouts. We have implemented a mitigation, and error rates are improving. We're monitoring the situation."

• Oct 16, 2026, 13:22 UTC · [GAPI] - [Global] - Qwen3 Embeddings produces timeouts

  30 minutes · rated "Minor"

  "The incident is now resolved. Customers should not see timeouts anymore. We continue to monitor the related deployments."

• Oct 6, 2026, 15:02 UTC · [Generative APIs] - [fr-par] - glm-5.2 service instabilities

  2 days 19 hours · rated "Minor"

  "The incident is now resolved, the situation is back to nomal."

September 2026

• Sep 23, 2026, 19:51 UTC · [fr-par] Qwen 3 235B saturated

  8 days 12 hours · rated "Minor"

  "[18:30] Qwen3 235b model performance is degraded, a few timeouts have been observed [21:30] Performance is still sub-optimal but timeouts are mostly gone"

• Sep 22, 2026, 14:57 UTC · [GAPI] - [GLOBAL] - gpt-oss-120b unresponsiveness

  duration not stated · marked as no impact

  "[10:34] During 3 minutes requests sent to the gpt-oss-120b could not be processed leading to 502 HTTP errors responses [10:37] Service back to normal. gpt-oss-120b able to process requests"

• Sep 15, 2026, 17:03 UTC · [GAPI] - Dedicated deployment cannot start properly

  13 hours 53 min · rated "Minor"

  "[17:20] A network issue causes an internal component to be unreachable, preventing the dedicated deployment to start properly."

• Sep 10, 2026, 10:06 UTC · [generative-apis] - [fr-par ] - gemma4 unhealthy

  18 days 3 hours · rated "Minor"

  "[09/09 - 16:50] The Gemma 4 models started to show un usual latencies resulting in requests timeout [10/09 - 9:00] The incident has been mitigated and response times are back to normal"