FUEL Status - Notice history

Third Party: Scaleway → Elements - Products → Private Network experiencing degraded performance

API - Operational

100% - uptime
May 2026 · 100%Jun · 99.6%Jul · 100%

Previous page

Next page

May 2026
Jun 2026
Jul 2026

Web Console - Operational

100% - uptime
May 2026 · 99.3%Jun · 99.7%Jul · 100%

Previous page

Next page

May 2026
Jun 2026
Jul 2026

MCP Server - Operational

100% - uptime
May 2026 · 100%Jun · 99.7%Jul · 100%

Previous page

Next page

May 2026
Jun 2026
Jul 2026

Billing - Operational

100% - uptime
May 2026 · 100%Jun · 99.7%Jul · 100%

Previous page

Next page

May 2026
Jun 2026
Jul 2026

Notice history

Jul 2026

Elevated error rates on the inference API
  • Resolved
    UTC
    Resolved
    This incident has been resolved.
  • Monitoring
    UTC
    Monitoring
    A fix has been deployed for the elevated error rates on the inference API. We are monitoring to confirm stability before marking the incident as resolved.
  • Identified
    UTC
    Identified
    We have identified the cause of elevated error rates on the inference API. A planned maintenance event by our infrastructure provider affected a node serving a subset of embedding model requests. We are working to restore full availability.
  • Investigating
    UTC
    Investigating
    We are investigating an elevated rate of errors affecting some inference API requests. Some customers may receive HTTP 503 responses when making requests. We are working to identify the cause and restore normal operation.
Elevated errors on SaaS Console analytics features
  • Resolved
    UTC
    Resolved
    This incident has been resolved.
  • Identified
    UTC
    Identified
    We are investigating elevated errors affecting the Usage, Monitoring, and Log tabs in the SaaS Console. Some customers may experience errors or blank data when viewing these sections. The team has identified the issue and is working on restoration.
Console — credit top-up and purchases unavailable
  • Resolved
    UTC
    Resolved
    This incident has been resolved.
  • Investigating
    UTC
    Investigating
    We are investigating an issue where wallet top-ups and credit purchases are unavailable in the console. Customers attempting to add credits may see errors. We will provide an update as soon as we have more information.
Elevated latency and timeouts on Deepseek V4 Flash
  • Resolved
    UTC
    Resolved
    This incident has been resolved.
  • Investigating
    UTC
    Investigating
    We are currently investigating this incident.

Jun 2026

Major Outage — All Services Unavailable
  • Resolved
    UTC
    Resolved

    This incident is resolved. The inference API, the console, the MCP endpoints, and documentation have been operating normally since recovery completed, confirmed stable across our monitoring window. Thank you for your patience.

  • Update
    UTC
    Update

    Embeddings are now restored. All services — both text generation and embeddings models, the console, the MCP endpoints, and documentation — are serving normally. We are continuing to monitor for stability before marking this incident resolved.

    Next update once we confirm sustained stability.

  • Monitoring
    UTC
    Monitoring

    Text generation is restored, and the console, MCP endpoints, and documentation are back online. Our embeddings model remains unavailable, so embeddings requests are still failing. We are working to restore it and monitoring the recovered services for stability.

    Next update within 30 minutes.

  • Identified
    UTC
    Identified

    The outage was caused by an internal infrastructure change that took our services offline. We have identified the cause and are restoring all affected services — the inference API, the console, the MCP endpoints, and documentation. Recovery is in progress.

    We're sorry for the disruption. Next update within 30 minutes.

  • Update
    UTC
    Update

    We are investigating a major outage affecting all customer-facing services. The inference API, the console, the MCP endpoints, and our documentation are currently unreachable, and requests are failing. Our engineers are actively working to restore service.

    Next update within 30 minutes.

  • Investigating
    UTC
    Investigating
    We are currently investigating this incident.
Major Outage — API and Console unavailable
  • Resolved
    UTC
    Resolved

    This incident is resolved. The API and Console have been fully operational and stable since 16:55 CEST. The earlier disruption was caused by an internal operational error that we've since corrected.

    We're sorry for the disruption and thank you for your patience.

  • Monitoring
    UTC
    Monitoring

    The API and Console are fully operational again, and we're monitoring for stability.

    We'll confirm full resolution once stability holds.

  • Update
    UTC
    Update

    The API is now serving most models normally. A subset of models is still being brought back online and may return errors in the meantime — we're actively restoring them.

    We'll post the next update by 17:30 CEST.

  • Update
    UTC
    Update

    The Console is back online and accessible. The API remains unavailable, and we are continuing to restore it as quickly as possible.

  • Identified
    UTC
    Identified

    All Fuel services — the API and the Console — are currently down. We've identified the cause and are working to restore service as quickly as possible.

    We're sorry for the disruption. We'll post the next update by 17:00 CEST.

  • Investigating
    UTC
    Investigating
    We are currently investigating this incident.
Scheduled GPU infrastructure maintenance
  • Completed
    June 03, 2026 at 8:00 PMUTC
    Completed
    June 03, 2026 at 8:00 PMUTC
    Maintenance has completed successfully
  • In progress
    June 03, 2026 at 7:30 PMUTC
    In progress
    June 03, 2026 at 7:30 PMUTC
    Maintenance is now in progress
  • Planned
    June 03, 2026 at 7:30 PMUTC
    Planned
    June 03, 2026 at 7:30 PMUTC

    We will be performing scheduled maintenance on our GPU infrastructure to improve platform reliability and performance. The maintenance will take approximately 30 minutes, during which model serving may experience a short interruption (up to ~10 minutes). Requests that fail during this period should succeed on retry. No action is required on your part, and no data is affected.

May 2026

Upstream Provider Incident — Scaleway fr-par
  • Resolved
    UTC
    Resolved

    Our provider has fully restored their Paris datacenter to nominal operation. We observed no customer-visible service disruption throughout the event. This incident is now resolved.

  • Update
    UTC
    Update

    Our provider's incident is nearing resolution, though their affected zone remains on backup power. We have not observed any customer-visible service disruption since our last update and continue to monitor closely.

  • Update
    UTC
    Update

    Our provider is still working to fully recover from the earlier incident in their Paris datacenter. On our side, we have not observed any further service disruption and are continuing to monitor closely.

    We'll post another update if the situation changes.

  • Monitoring
    UTC
    Monitoring

    Most services have recovered. Scaleway identified a power issue in their Paris datacenter and is working to restore the rest. We'll update again within 30 minutes, or sooner once fully resolved.

  • Identified
    UTC
    Identified

    We are currently experiencing a full service outage. All our services are down due to an ongoing networking incident at our infrastructure provider, Scaleway (Paris region).

    We are monitoring Scaleway's investigation closely and will restore service as soon as upstream connectivity is recovered. No action is required on your side.

    Upstream incident: https://status.scaleway.com/incidents/qkj49g6g3ykp

    We'll post the next update within 30 minutes, or sooner if the situation changes. We sincerely apologize for the disruption.

[Retroactive] Intermittent "Access denied" errors on console.fuel1.ai for some users
  • Resolved
    UTC
    Resolved

    The issue has been fully resolved and we've verified that customer portal sign-in is working consistently for all users. We're following up with our identity provider on the underlying cause and improving our monitoring to detect any recurrence quickly.

  • Identified
    UTC
    Identified

    We've identified the cause as an intermittent issue with our authentication provider that, for some users, prevented their organization membership from being recognised at sign-in. Only sign-in via the customer portal at console.fuel1.ai was impacted; API access using API keys was not affected. We're working with our identity provider on remediation. No customer data was lost or exposed.

  • Investigating
    UTC
    Investigating

    We're investigating reports of users seeing "Access denied" on console.fuel1.ai despite being correctly added to an organization. Inference API access via API keys is unaffected.

May 2026 to Jul 2026

Next