Nebius AI Studio outage history

Biggest Nebius AI Studio 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. Sep 14, 2026: Managed Service for SkyPilot: cluster creation and update requests failing in eu-north1 24 hours 33 min · rated "Major"
2. Aug 19, 2026: Region us-central1 partial degradation 21 hours 33 min · rated "Major"
3. May 6, 2026: Managed Kubernetes operations can take longer than expected 19 hours 6 min · rated "Major"
4. Apr 29, 2026: Partial unavailability of external models in Token Factory 16 hours 29 min · rated "Major"
5. Oct 7, 2026: Problems with power for several dozens of nodes 14 hours 45 min · rated "Major"
6. Oct 15, 2026: InfiniBand connectivity issues for GPU clusters in eu-west2 10 hours 38 min · rated "Major"
7. Sep 16, 2026: [us-central1] Object Storage is partially unavailable 4 hours 57 min · rated "Major"
8. Aug 15, 2026: Partial InfiniBand unavailability 3 hours 17 min · rated "Major"
9. Sep 23, 2026: Network issues in eu-west2 3 hours 12 min · rated "Major"

Every smaller Nebius AI Studio incident

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

October 2026

• Oct 18, 2026, 14:07 UTC · Partial IAM and console unavailability

  duration not stated · marked as no impact

  "The console and IAM were unavailable for approximately 10 minutes. The incident has been mitigated, and service availability has been restored."

• Oct 16, 2026, 13:02 UTC · CLI login issues

  11 minutes · rated "Major"

  "Some users may experience login failures when authenticating with the Nebius CLI via `auth.eu.nebius.com`."

• Oct 14, 2026, 20:41 UTC · Connectivity issues between eu-south1 and other Nebius regions

  9 hours 9 min · rated "Minor"

  "We have identified the root cause of the connectivity issues and are preparing a fix. Currently, disk snapshot creation in NBS may be affected. Additionally, Nebius Console and API may experience higher latency than usual."

• Oct 14, 2026, 09:54 UTC · InfiniBand connectivity issues for GPU clusters in eu-west2

  58 minutes · rated "Major"

  "We have identified the cause of the InfiniBand connectivity issues. Our on-site team is working to resolve it. We will provide an update as soon as we have more information."

• Oct 8, 2026, 08:48 UTC · Metrics unavailable in Management Console for Monitoring in eu-west1

  11 minutes · rated "Minor"

  "The incident has been resolved."

• Oct 4, 2026, 15:54 UTC · Issues with preempted Managed Kubernetes nodes in all regions

  24 hours · rated "Minor"

  "The issue has been identified and a fix has been prepared. The issue has limited impact and does not affect running workloads."

• Oct 3, 2026, 02:30 UTC · InfiniBand connectivity issues for GPU clusters in eu-west2

  6 hours 28 min · rated "Minor"

  "We have identified the cause of the InfiniBand connectivity issues. Our on-site team is working to resolve it. We will provide an update as soon as we have more information."

• Oct 2, 2026, 01:58 UTC · Compute VM creation and Object storage are partial unavailable

  9 minutes · rated "Major"

  "Object Storage was partially unavailable in eu-west1. Moreover, we observed problems around creation of new virtual machines and disks in several regions. Now problems have been fixed and teams are monitoring the situation"

September 2026

• Sep 22, 2026, 09:21 UTC · Object storage unavailable in eu-west1

  duration not stated · rated "Major"

  "Incident is resolved."

• Sep 17, 2026, 16:26 UTC · Compute services are partially unavailable for customers

  43 minutes · rated "Major"

  "The root cause was found and this incident has been resolved."

• Sep 17, 2026, 10:46 UTC · Resuming IAM accounts is unavailable

  58 minutes · marked as no impact

  "We are investigating an issue with IAM Tenants Status change. Customers may not be able to resume their accounts from the suspended state."

• Sep 16, 2026, 16:52 UTC · [eu-west1] Metrics for resources in this region are partially unavailable

  1 hour 29 min · rated "Major"

  "This incident has been resolved."

• Sep 9, 2026, 12:17 UTC · Unstable compute VM create

  24 minutes · rated "Minor"

  "In some cases VMs may fail to create or metadata inside may not be available. The team has identified the issue and is working on deploying the fix."

• Sep 1, 2026, 16:11 UTC · Connectivity issues for newly created worker nodes

  15 minutes · rated "Major"

  "Worker nodes of Managed Service for Kubernetes clusters created after 15:13 CET Sep 1 2026 could experience problems connecting to/from other nodes in the same cluster. The root cause is understood and mitigation is underway."

August 2026

• Aug 17, 2026, 13:00 UTC · Issues logging to the Management Console

  duration not stated · rated "Minor"

  "Some newly registered users may have experienced issues logging in to the management console between 13:00 UTC on 17 August 2026 and 14:00 UTC on 18 August 2026."

• Aug 12, 2026, 13:30 UTC · External VPC connectivity issues

  26 minutes · rated "Major"

  "We are investigating an issue with external VPC connectivity"

• Aug 5, 2026, 14:27 UTC · Creation of new VMs is broken

  1 hour 18 min · marked as no impact

  "Due to issue with internal services the creation of new VM and other operations may be affected. We are investigating the issue."

July 2026

• Jul 27, 2026, 18:30 UTC · Network issues

  1 hour 33 min · rated "Major"

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

• Jul 22, 2026, 15:00 UTC · [eu-north1] Network timeouts affecting multiple services

  duration not stated · rated "Major"

  "The eu-north1 region experienced network issues that may have affected the operation of several services. The issue has been resolved, and all services are operating normally."

• Jul 14, 2026, 22:10 UTC · InfiniBand switch failure affecting GPU workloads in the eu-north1-c fabric in the eu-north1 region

  22 minutes · rated "Major"

  "This incident has been resolved."

• Jul 3, 2026, 14:50 UTC · Managed SkyPilot OIDC Login is unavailable

  duration not stated · rated "Minor"

  "Managed SkyPilot OIDC Login was unavailable since 2 of July. The problem has been identified and resolved."

• Jul 1, 2026, 22:33 UTC · Monitoring, Object Storage and Token Factory services are partially unavailable

  1 hour 44 min · rated "Major"

  "We are currently experiencing partial unavailability of our Monitoring, Object Storage, Token Factory services. Our team is actively investigating the issue and working towards a resolution."

June 2026

• Jun 30, 2026, 13:24 UTC · New virtual machines are not being created

  4 minutes · rated "Minor"

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

• Jun 24, 2026, 14:08 UTC · Object Storage service partially unavailable in US-CENTRAL1 region

  1 hour 37 min · rated "Major"

  "The Object Storage service in the US-CENTRAL1 region has been fully restored and is operating normally. The root cause has been identified, and corrective measures are in place."

• Jun 16, 2026, 20:02 UTC · One InfiniBand switch failure

  32 minutes · rated "Minor"

  "One InfiniBand switch in the fabric-3 cluster is down."

• Jun 15, 2026, 18:27 UTC · Network Connectivity Issues

  21 minutes · rated "Minor"

  "Connectivity issues are gone now. Intermittent failure is being investigated in the background with the uplink provider, while the Nebius Netowrking team continues to monitor the link closely."

• Jun 10, 2026, 20:56 UTC · Compute instances startup issues

  2 hours 33 min · marked as no impact

  "We're currently investigating an issue affecting Compute instances startup. Some requests to create or start instances may fail. Already-running instances are not affected. Our engineering team is actively investigating."

• Jun 9, 2026, 13:25 UTC · Some public IPs unreachable in eu-north1

  duration not stated · rated "Minor"

  "Some of public IPs became unreachable."

• Jun 4, 2026, 09:46 UTC · Network Connectivity Issues

  1 hour 28 min · rated "Major"

  "An issue with Internet connectivity in the data center has been detected. Impact: - Logs and metrics from virtual machines are not being delivered. - Object Storage is currently unavailable from internet."

• Jun 2, 2026, 10:53 UTC · Unable to allocate new public IPv4 addresses in eu-north1

  24 hours 13 min · rated "Minor"

  "We have identified the cause: the pool of available public IPv4 addresses in eu-north1 is currently exhausted, which is why new public IPv4 addresses cannot be allocated to workloads."

May 2026

• May 29, 2026, 22:20 UTC · Power issues

  duration not stated · marked as no impact

  "Several servers were rebooted."

• May 19, 2026, 13:45 UTC · Unable to perform operations on a small subset of virtual machines

  1 hour 55 min · rated "Minor"

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

• May 13, 2026, 18:45 UTC · InfiniBand performance degradation

  2 hours 20 min · rated "Minor"

  "We are investigating reduced InfiniBand fabric performance affecting GPU workloads in the uk-south1 region. Customers may observe lower throughput and slower distributed training jobs."

• May 7, 2026, 21:03 UTC · Nebius console is unavailable

  24 minutes · marked as no impact

  "Issue resolved 21:10."

April 2026

• Apr 24, 2026, 09:46 UTC · Partial degradation of Dedicated Endpoints in Token Factory

  43 minutes · rated "Minor"

  "A fix has been implemented and we are monitoring the results."

• Apr 20, 2026, 08:05 UTC · Metrics for VMs are unavailable

  25 minutes · rated "Minor"

  "Since 8:30 back to normal."

• Apr 20, 2026, 07:26 UTC · TokenFactory DataLab unavailability

  25 hours 49 min · rated "Minor"

  "We are continuing to investigate this issue."

• Apr 15, 2026, 08:50 UTC · TokenFactory - Some Serverless Models Partially Unavailable in us-central1 Region

  1 hour 36 min · rated "Minor"

  "Users may experience issues when working with the following serverless models in the us-central1 region: - Qwen/Qwen3.5-397B-A17B - MiniMaxAI/MiniMax-M2."

• Apr 10, 2026, 02:00 UTC · Degraded logs, metrics and audit trails availability

  duration not stated · rated "Minor"

  "During the night we experienced issues with storage service. While writing for data was not affected (just delayed), but reading logs, accessing metrics and audit trails in console was not possible while the incident was in active stage."

• Apr 7, 2026, 11:50 UTC · Observability data partially unavailable for Dedicated Endpoints in TokenFactory

  1 hour 14 min · rated "Minor"

  "We are noticing issues with some customers viewing metrics for Dedicated Endpoints in Token Factory. We are currently investigating this issue."

• Apr 1, 2026, 13:58 UTC · Slower project and tenant creation

  60 minutes · rated "Minor"

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

March 2026

• Mar 26, 2026, 10:20 UTC · Managed k8s is partially unavailable for customers

  36 minutes · rated "Major"

  "A fix has been implemented and we are monitoring the results"

• Mar 24, 2026, 23:30 UTC · Storage Metrics Ingestion Failure

  duration not stated · rated "Major"

  "We had a problem with metrics ingestion which resulted in storage metrics not being ingested during the following intervals: - eu-west: from 14:50 CET to 19:05 CET - me-west1: from 17:00 CET to 19:05 CET - eu-west1: from 15:50 CET till..."