Skip to content
UpButler
ProductAgentsStatus pagesDocsPricing
Sign inStart free

Is Nebius down?

Nebius is up

Right now

No. No incidents reported.

Last checked (2026-10-10 07:17 UTC) from status.nebius.com · every 180s

For agents

proceed

Proceed.

GET upbutler.com/status/nebius/status.agent.json

MCP: upstream_status {"service":"nebius"}

AI cloud GPUs and the Nebius AI Studio inference API.

What Nebius says

Operational

Taken as is from Nebius's own status page API. Incident titles and component names below are theirs.

Open status.nebius.com

What UpButler measured

There is no keyless endpoint of Nebius that says anything useful, so UpButler does not probe it. To check your own Nebius account, add a monitor with your key: it is yours alone and catches what a shared probe cannot.

Nebius components

80 operational

EU-NORTH1

  • Compute CloudOperational
  • Network Block StorageOperational
  • Object StorageOperational
  • Virtual Private Cloud (Networking)Operational
  • Managed Service for Kubernetes®Operational
  • Identity and Access ManagementOperational
  • BillingOperational
  • APIOperational
  • Management ConsoleOperational

EU-NORTH2

  • Compute CloudOperational
  • Network Block StorageOperational
  • Object StorageOperational
  • Virtual Private Cloud (Networking)Operational
  • Managed Service for Kubernetes®Operational
  • Identity and Access ManagementOperational
  • BillingOperational
  • APIOperational
  • Management ConsoleOperational

EU-WEST1

  • Compute CloudOperational
  • Network Block StorageOperational
  • Object StorageOperational
  • Virtual Private Cloud (Networking)Operational
  • Managed Service for Kubernetes®Operational
  • Identity and Access ManagementOperational
  • BillingOperational
  • APIOperational
  • Management ConsoleOperational

US-CENTRAL1

  • Compute CloudOperational
  • Network Block StorageOperational
  • Object StorageOperational
  • Virtual Private Cloud (Networking)Operational
  • Managed Service for Kubernetes®Operational
  • Identity and Access ManagementOperational
  • BillingOperational
  • APIOperational
  • Management ConsoleOperational

ME-WEST1

  • Compute CloudOperational
  • Network Block StorageOperational
  • Object StorageOperational
  • Virtual Private Cloud (Networking)Operational
  • Managed Service for Kubernetes®Operational
  • Identity and Access ManagementOperational
  • BillingOperational
  • APIOperational
  • Management ConsoleOperational

UK-SOUTH1

  • Compute CloudOperational
  • Network Block StorageOperational
  • Object StorageOperational
  • Virtual Private Cloud (Networking)Operational
  • Managed Service for Kubernetes®Operational
  • Identity and Access ManagementOperational
  • BillingOperational
  • APIOperational
  • Management ConsoleOperational

EU-WEST2

  • Compute CloudOperational
  • Network Block StorageOperational
  • Object StorageOperational
  • Virtual Private Cloud (Networking)Operational
  • Managed Service for Kubernetes®Operational
  • Identity and Access ManagementOperational
  • BillingOperational
  • APIOperational
  • Management ConsoleOperational

US-NORTH1

  • Compute CloudOperational
  • Network Block StorageOperational
  • Object StorageOperational
  • Virtual Private Cloud (Networking)Operational
  • Managed Service for Kubernetes®Operational
  • Identity and Access ManagementOperational
  • BillingOperational
  • APIOperational
  • Management ConsoleOperational

EU-SOUTH1

  • Compute CloudOperational
  • Network Block StorageOperational
  • Object StorageOperational
  • Virtual Private Cloud (Networking)Operational
  • Managed Service for Kubernetes®Operational
  • Identity and Access ManagementOperational
  • BillingOperational
  • APIOperational

Nebius incidents, last 90 days

  1. 2026-07-13: no record
  2. 2026-07-14: no record
  3. 2026-07-15: no record
  4. 2026-07-16: no record
  5. 2026-07-17: no record
  6. 2026-07-18: no record
  7. 2026-07-19: no record
  8. 2026-07-20: no record
  9. 2026-07-21: no record
  10. 2026-07-22: no record
  11. 2026-07-23: no record
  12. 2026-07-24: no record
  13. 2026-07-25: no record
  14. 2026-07-26: no record
  15. 2026-07-27: no record
  16. 2026-07-28: no record
  17. 2026-07-29: no record
  18. 2026-07-30: no record
  19. 2026-07-31: no record
  20. 2026-08-01: no record
  21. 2026-08-02: no record
  22. 2026-08-03: no record
  23. 2026-08-04: no record
  24. 2026-08-05: no record
  25. 2026-08-06: no record
  26. 2026-08-07: no record
  27. 2026-08-08: no record
  28. 2026-08-09: no record
  29. 2026-08-10: no record
  30. 2026-08-11: no record
  31. 2026-08-12: no record
  32. 2026-08-13: no record
  33. 2026-08-14: no record
  34. 2026-08-15: no record
  35. 2026-08-16: no record
  36. 2026-08-17: no record
  37. 2026-08-18: no record
  38. 2026-08-19: no record
  39. 2026-08-20: no record
  40. 2026-08-21: no record
  41. 2026-08-22: no record
  42. 2026-08-23: no record
  43. 2026-08-24: no record
  44. 2026-08-25: no record
  45. 2026-08-26: no record
  46. 2026-08-27: no record
  47. 2026-08-28: no record
  48. 2026-08-29: no record
  49. 2026-08-30: no record
  50. 2026-08-31: no record
  51. 2026-09-01: no record
  52. 2026-09-02: no record
  53. 2026-09-03: no record
  54. 2026-09-04: no record
  55. 2026-09-05: no record
  56. 2026-09-06: no record
  57. 2026-09-07: no record
  58. 2026-09-08: no record
  59. 2026-09-09: no record
  60. 2026-09-10: no record
  61. 2026-09-11: no record
  62. 2026-09-12: no record
  63. 2026-09-13: no record
  64. 2026-09-14: no record
  65. 2026-09-15: no record
  66. 2026-09-16: no record
  67. 2026-09-17: no record
  68. 2026-09-18: no record
  69. 2026-09-19: no record
  70. 2026-09-20: no record
  71. 2026-09-21: no record
  72. 2026-09-22: no record
  73. 2026-09-23: no record
  74. 2026-09-24: no record
  75. 2026-09-25: no record
  76. 2026-09-26: no record
  77. 2026-09-27: no record
  78. 2026-09-28: no record
  79. 2026-09-29: no record
  80. 2026-09-30: no record
  81. 2026-10-01: no record
  82. 2026-10-02: major incident
  83. 2026-10-03: minor incident
  84. 2026-10-04: minor incident
  85. 2026-10-05: minor incident
  86. 2026-10-06: no incident reported
  87. 2026-10-07: major incident
  88. 2026-10-08: major incident
  89. 2026-10-09: no incident reported
  90. 2026-10-10: no incident reported
2026-07-136 of 9 days with an incidenttoday

Each bar is a day, coloured by the worst incident Nebius reported. It is their incident record, not a measurement of uptime.

Incidents
5
2 major or critical
Typical time to resolve
6 h 28 min
median
Nine in ten within
24 h
90th percentile
Longest
24 h
record since 2026-10-02
Started (UTC)IncidentImpactLasted
Metrics unavailable in Management Console for Monitoring in eu-west1minor11 min
Problems with power for several dozens of nodesmajor14 h 45 min
Issues with preempted Managed Kubernetes nodes in all regionsminor24 h
InfiniBand connectivity issues for GPU clusters in eu-west2minor6 h 28 min
Compute VM creation and Object storage are partial unavailablemajor9 min

Get alerted when Nebius goes down

No account. One email when Nebius reports an incident, one when it is over. You confirm the address first, and every email has a one-click unsubscribe.

Or follow the RSS feed. On UpButler? Add Nebius as a dependency to route it to Slack, Teams or a webhook.

Status badge

Show Nebius's current status in a README or an internal wiki. It updates by itself.

Nebius status badge

[![Nebius status](https://upbutler.com/status/nebius/badge.svg)](https://upbutler.com/status/nebius)

Tell your outage from Nebius's

Declare that your project depends on Nebius. When one of your monitors fails while Nebius has an incident that began around the same time, UpButler tags your incident "likely upstream" with their incident linked, says so in the alert and the AI report, and tells your on-call agent not to change code.

npx upbutler init finds the dependency from your env var names and packages. The catalog, dependencies and upstream-aware incidents are on every plan, including Free.

Start freeHow it works
upbutler.yaml
# upbutler.yaml
dependencies:
  - nebius
Ask before you retry
curl -s https://upbutler.com/status/nebius/status.agent.json
status.agent.json (live)
{
  "schema": "upbutler.upstream-status/v1",
  "service": "nebius",
  "verdict": "proceed",
  "status": "operational",
  "reason": "No incidents reported",
  "retryAfterSec": null,
  "incident": null,
  "probe": null
}

Questions

Is Nebius down right now?
No. As of 2026-10-10 07:17 UTC, Nebius is up. This page reads the status again every 3 minutes.
How long do Nebius outages usually last?
Of the 5 incidents Nebius reported in the last 90 days, half were resolved within 6 h 28 min and nine in ten within 24 h; the longest took 24 h. These are the provider's own start and end times.
Where does this status come from?
From Nebius's own status page API at https://status.nebius.com. UpButler reads it once for everyone (not once per visitor) every 3 minutes with conditional requests, and shows it unchanged: incident titles and component names are Nebius's words.
Can I get an alert when Nebius goes down?
Yes, without an account: enter your email on this page, confirm the link we send, and you get one email when Nebius reports an incident and one when it is over. There is also an RSS feed at https://upbutler.com/status/nebius/feed.rss. Teams on UpButler can route the same alert to Slack, Teams or a webhook.
How can my agent or app check Nebius status?
Fetch https://upbutler.com/status/nebius/status.agent.json: a small JSON document with a verdict (proceed, retry, fallback or pause) and retryAfterSec. No key is needed. The same answer is an MCP tool, upstream_status {"service":"nebius"}, on https://upbutler.com/mcp/public.
How do I know whether an outage is mine or Nebius's?
Add "nebius" to the dependencies of your UpButler workspace. When one of your monitors fails while Nebius has an incident that began around the same time, your incident is tagged "likely upstream" with their incident linked, and your on-call agent is told not to change code.

Also check

UpButler is not affiliated with Nebius. Status and incident text belong to Nebius and are read from their public status source; the probe result is UpButler's own measurement.