Official provider signal

Is Vercel down?

Vercel builds, deployments, edge network and the dashboard.

Operational

Current status

Components

38

Builds
Operational
Blob
Operational
Output Generation
Operational
AI Gateway
Operational
Dashboard
Operational
Drains
Operational
Builds
Operational
Global Config
Operational
Dashboard
Operational
API
Operational
Support Center
Operational
Logs
Operational
Observability
Operational
Deploy Hooks
Operational
Support Chat
Operational
Speed Insights
Operational
Git Integrations
Operational
Web Analytics
Operational
Community
Operational
Cron Jobs
Operational
Integration Webhooks
Operational
Data Cache
Operational
DNS
Operational
Domain Registration
Operational
Firewall
Operational
Functions
Operational
Image Optimization
Operational
Marketplace
Operational
Queues
Operational
Remote Caching
Operational
Routing Middleware
Operational
SAML Single Sign-On
Operational
Sandbox
Operational
TLS Certificates
Operational
Workflow
Operational
Container Registry
Operational
Key Management Service
Operational
Connect
Operational

Official timeline

Active incidents

incidentDeployment stuck in initializing state
monitoring
Timeline →
incidentElevated Errors Triggering Deployments
monitoring
Timeline →
incidentElevated error rate on Connect and Passport
identified
Timeline →
incidentElevated latency in VCR, Blob, and Sandbox API
monitoring
Timeline →

Data confidence

Source & freshness

Source: Vercel Status ↗
Collected: Sep 23, 2026, 6:37 PM UTC
Provider updated: Sep 23, 2026, 6:36 PM UTC

Operational means no issue is currently reported by the provider. It is not proof that every user or region is unaffected.

While this is open

What should I check next?

Read the incident timeline. The provider updates say what is affected and what they have already ruled out.
Match the affected components against the calls your system actually makes before assuming this incident is yours.
Hold off on rolling back your own deploy. A rollback during a provider incident removes a variable you will want later.
Queue or fail over the work if you can. Aggressive retries against a degraded provider usually make the backlog worse.
Watch the timeline rather than polling the API. It updates as the provider posts, including the resolution.