Is Weights & Biases down right now?

No. As of 2026-10-10 07:18 UTC, Weights & Biases is up. This page reads the status again every 5 minutes.

How long do Weights & Biases outages usually last?

Of the 5 incidents Weights & Biases reported in the last 90 days, half were resolved within 77 days and nine in ten within 80 days; the longest took 80 days. These are the provider's own start and end times.

Where does this status come from?

From Weights & Biases's own incident feed at https://status.wandb.com. UpButler reads it once for everyone (not once per visitor) every 5 minutes with conditional requests, and shows it unchanged: incident titles and component names are Weights & Biases's words.

Can I get an alert when Weights & Biases goes down?

Yes, without an account: enter your email on this page, confirm the link we send, and you get one email when Weights & Biases reports an incident and one when it is over. There is also an RSS feed at https://upbutler.com/status/wandb/feed.rss. Teams on UpButler can route the same alert to Slack, Teams or a webhook.

How can my agent or app check Weights & Biases status?

Fetch https://upbutler.com/status/wandb/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":"wandb"}, on https://upbutler.com/mcp/public.

How do I know whether an outage is mine or Weights & Biases's?

Add "wandb" to the dependencies of your UpButler workspace. When one of your monitors fails while Weights & Biases 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.

Is Weights & Biases down?

Weights & Biases is up

Right now

No. No incidents reported.

Last checked 64 s ago (2026-10-10 07:18 UTC) from status.wandb.com · every 300s

For agents

proceed

Proceed.

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

MCP: upstream_status {"service":"wandb"}

Experiment tracking, Weave tracing and model registry.

What Weights & Biases says

Operational

Taken as is from Weights & Biases's own incident feed. Incident titles and component names below are theirs.

Open status.wandb.com

What UpButler measured

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

Weights & Biases incidents, last 90 days

2026-07-13: minor incident
2026-07-14: minor incident
2026-07-15: minor incident
2026-07-16: minor incident
2026-07-17: minor incident
2026-07-18: minor incident
2026-07-19: minor incident
2026-07-20: minor incident
2026-07-21: minor incident
2026-07-22: minor incident
2026-07-23: minor incident
2026-07-24: minor incident
2026-07-25: minor incident
2026-07-26: minor incident
2026-07-27: minor incident
2026-07-28: minor incident
2026-07-29: minor incident
2026-07-30: minor incident
2026-07-31: minor incident
2026-08-01: minor incident
2026-08-02: minor incident
2026-08-03: minor incident
2026-08-04: minor incident
2026-08-05: minor incident
2026-08-06: minor incident
2026-08-07: minor incident
2026-08-08: minor incident
2026-08-09: minor incident
2026-08-10: minor incident
2026-08-11: minor incident
2026-08-12: minor incident
2026-08-13: minor incident
2026-08-14: minor incident
2026-08-15: minor incident
2026-08-16: minor incident
2026-08-17: minor incident
2026-08-18: minor incident
2026-08-19: minor incident
2026-08-20: minor incident
2026-08-21: minor incident
2026-08-22: minor incident
2026-08-23: minor incident
2026-08-24: minor incident
2026-08-25: minor incident
2026-08-26: minor incident
2026-08-27: minor incident
2026-08-28: minor incident
2026-08-29: minor incident
2026-08-30: minor incident
2026-08-31: minor incident
2026-09-01: major incident
2026-09-02: major incident
2026-09-03: major incident
2026-09-04: major incident
2026-09-05: major incident
2026-09-06: major incident
2026-09-07: major incident
2026-09-08: major incident
2026-09-09: major incident
2026-09-10: major incident
2026-09-11: major incident
2026-09-12: major incident
2026-09-13: major incident
2026-09-14: major incident
2026-09-15: major incident
2026-09-16: major incident
2026-09-17: major incident
2026-09-18: major incident
2026-09-19: major incident
2026-09-20: major incident
2026-09-21: major incident
2026-09-22: major incident
2026-09-23: major incident
2026-09-24: major incident
2026-09-25: major incident
2026-09-26: major incident
2026-09-27: major incident
2026-09-28: major incident
2026-09-29: major incident
2026-09-30: major incident
2026-10-01: no incident reported
2026-10-02: no incident reported
2026-10-03: no incident reported
2026-10-04: no incident reported
2026-10-05: no incident reported
2026-10-06: no incident reported
2026-10-07: no incident reported
2026-10-08: no incident reported
2026-10-09: no incident reported
2026-10-10: no incident reported

2026-07-1380 of 90 days with an incidenttoday

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

Incidents: 5; 1 major or critical
Typical time to resolve: 77 days; median
Nine in ten within: 80 days; 90th percentile
Longest: 80 days; record since 2026-07-12

Started (UTC)	Incident	Impact	Lasted
2026-09-01 14:40	API Degraded	major	29 days
2026-07-16 00:06	Errors fetching media files	minor	77 days
2026-07-15 06:48	Metric Ingestion Backup	minor	77 days
2026-07-13 19:35	Metrics Ingestion Backup	minor	79 days
2026-07-12 17:52	Metric Ingestion Backlog	minor	80 days

Get alerted when Weights & Biases goes down

No account. One email when Weights & Biases 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 Weights & Biases as a dependency to route it to Slack, Teams or a webhook.

Status badge

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

[![Weights & Biases status](https://upbutler.com/status/wandb/badge.svg)](https://upbutler.com/status/wandb)

Tell your outage from Weights & Biases's

Declare that your project depends on Weights & Biases. When one of your monitors fails while Weights & Biases 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 free How it works

upbutler.yaml

# upbutler.yaml
dependencies:
  - wandb

Ask before you retry

curl -s https://upbutler.com/status/wandb/status.agent.json

status.agent.json (live)

{
  "schema": "upbutler.upstream-status/v1",
  "service": "wandb",
  "verdict": "proceed",
  "status": "operational",
  "reason": "No incidents reported",
  "retryAfterSec": null,
  "incident": null,
  "probe": null
}

Questions

Is Weights & Biases down right now?: No. As of 2026-10-10 07:18 UTC, Weights & Biases is up. This page reads the status again every 5 minutes.
How long do Weights & Biases outages usually last?: Of the 5 incidents Weights & Biases reported in the last 90 days, half were resolved within 77 days and nine in ten within 80 days; the longest took 80 days. These are the provider's own start and end times.
Where does this status come from?: From Weights & Biases's own incident feed at https://status.wandb.com. UpButler reads it once for everyone (not once per visitor) every 5 minutes with conditional requests, and shows it unchanged: incident titles and component names are Weights & Biases's words.
Can I get an alert when Weights & Biases goes down?: Yes, without an account: enter your email on this page, confirm the link we send, and you get one email when Weights & Biases reports an incident and one when it is over. There is also an RSS feed at https://upbutler.com/status/wandb/feed.rss. Teams on UpButler can route the same alert to Slack, Teams or a webhook.
How can my agent or app check Weights & Biases status?: Fetch https://upbutler.com/status/wandb/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":"wandb"}, on https://upbutler.com/mcp/public.
How do I know whether an outage is mine or Weights & Biases's?: Add "wandb" to the dependencies of your UpButler workspace. When one of your monitors fails while Weights & Biases 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 Weights & Biases. Status and incident text belong to Weights & Biases and are read from their public status source; the probe result is UpButler's own measurement.