Some services are down
Affected: Managed Endpoints (Endpoints - Kimi K3)
Timeline · 1 update
- investigating Aug 03, 2026, 06:53 PM UTC
Endpoints - Kimi K3 went down.
Modal Outage History
Modal is up right nowModal had 47 outages in the last 2 years totaling 11h 52m of downtime — averaging 1.9 incidents per month.
There were 47 Modal outages since December 28, 2025 totaling 11h 52m of downtime. Each is summarised below — incident details, duration, and resolution information.
Web endpoints recovered
Timeline · 1 update
- resolved Jul 30, 2026, 03:47 PM UTC
Web endpoints recovered
Flash Proxy (us-west) recovered
Timeline · 1 update
- resolved Jul 24, 2026, 11:13 AM UTC
Flash Proxy (us-west) recovered
Flash Proxy (us-west) is down
Affected: Flash Proxy (Flash Proxy (us-west))
Timeline · 2 updates
- investigating Jul 24, 2026, 11:11 AM UTC
Flash Proxy (us-west) went down.
- resolved Jul 24, 2026, 11:15 AM UTC
Flash Proxy (us-west) recovered.
modal.com is degraded
Affected: Web services (Frontend)
Timeline · 2 updates
- investigating Jul 23, 2026, 04:19 AM UTC
We are investigating elevated rates of 5xx errors on modal.com
- resolved Jul 23, 2026, 04:50 AM UTC
We identified the cause and deployed a fix and seeing recovery.
Storage degraded - impactin...
Affected: Storage (Volumes)
Timeline · 1 update
- investigating Jun 29, 2026, 09:02 PM UTC
We're tracking errors in our storage subsystem which is causing issues with volumes and app deploys.
Image builds recovered
Timeline · 1 update
- resolved Jun 16, 2026, 06:56 PM UTC
Image builds recovered
Web API recovered
Timeline · 1 update
- resolved Jun 16, 2026, 05:18 PM UTC
Web API recovered
Image builds are down
Affected: Image builds (Image builds)
Timeline · 3 updates
- investigating Jun 16, 2026, 05:18 PM UTC
Image builds are failing intermittently, we're working on mitigation
- investigating Jun 16, 2026, 05:18 PM UTC
New Image builds are now processing normally. We have a backlog of old Image builds we are burning down.
- resolved Jun 16, 2026, 07:00 PM UTC
Image builds have recovered and caught up
Container scheduling / app...
Affected: Function execution (CPU functions)Function execution (GPU functions)Function execution (Web endpoints)Web services (Frontend)Web services (Web API)Sandbox execution (Sandboxes)Image builds (Image builds)Flash Proxy (Flash Proxy (us-west))
Timeline · 2 updates
- investigating Jun 16, 2026, 04:52 PM UTC
Container scheduling and some misc operations (e.g. app deploys) are down. We are actively working on mitigating.
- resolved Jun 16, 2026, 05:18 PM UTC
We've recovered operation
CLI volume operations disru...
Affected: Storage (Volumes)
Timeline · 2 updates
- investigating Jun 16, 2026, 03:20 PM UTC
Certain operations with the Modal CLI, such as modal volume put, will fail for files larger than a certain size.
- investigating Jun 16, 2026, 03:20 PM UTC
Volume CLI operations are no longer degraded.
Web endpoints is down
Affected: Function execution (Web endpoints)
Timeline · 1 update
- investigating Jun 13, 2026, 07:20 PM UTC
Web endpoints went down.
Web endpoints and Flash Pro...
Affected: Function execution (Web endpoints)Flash Proxy (Flash Proxy (us-west))
Timeline · 1 update
- investigating Jun 12, 2026, 08:44 PM UTC
Web endpoints went down.
Snapshot restores (beta) recovered
Timeline · 1 update
- resolved Jun 03, 2026, 02:54 PM UTC
Snapshot restores (beta) recovered
CPU functions recovered
Timeline · 1 update
- resolved Jun 03, 2026, 02:04 PM UTC
CPU functions recovered
Volumes recovered
Timeline · 1 update
- resolved Jun 03, 2026, 02:04 PM UTC
Volumes recovered
GPU functions recovered
Timeline · 1 update
- resolved Jun 03, 2026, 01:51 PM UTC
GPU functions recovered
Web endpoints, Volumes, and...
Affected: Function execution (Web endpoints)Function execution (Snapshot restores (beta))Storage (Volumes)Image builds (Image builds)
Timeline · 1 update
- investigating Jun 03, 2026, 01:05 PM UTC
Web endpoints went down.
auth subsystem failures cau...
Affected: Function execution (CPU functions)Function execution (GPU functions)Function execution (Web endpoints)Function execution (Snapshot restores (beta))Web services (Frontend)Web services (Web API)Storage (Volumes)Sandbox execution (Sandboxes)Image builds (Image builds)
Timeline · 2 updates
- investigating Jun 03, 2026, 01:03 PM UTC
A failure in our auth subsystem is causing failures affecting several services.
- resolved Jun 03, 2026, 01:52 PM UTC
Seeing recovery.
Database incident affecting...
Affected: Sandbox execution (Sandboxes)
Timeline · 2 updates
- investigating Jun 01, 2026, 04:39 PM UTC
Sandboxes are not starting due to a suspected database issue.
- resolved Jun 01, 2026, 04:52 PM UTC
Sandboxes have recovered. We continue to investigate the root cause.
Web endpoints and Image bui...
Affected: Function execution (Web endpoints)Image builds (Image builds)
Timeline · 3 updates
- investigating Jun 01, 2026, 03:40 PM UTC
Web endpoints went down.
- resolved Jun 01, 2026, 03:42 PM UTC
Web endpoints recovered.
- resolved Jun 01, 2026, 04:53 PM UTC
Image builds recovered.
CPU functions, GPU function...
Affected: Function execution (CPU functions)Function execution (GPU functions)Function execution (Snapshot restores (beta))Web services (Frontend)Storage (Volumes)Sandbox execution (Sandboxes)Image builds (Image builds)Flash Proxy (Flash Proxy (us-west))
Timeline · 2 updates
- investigating May 19, 2026, 01:29 PM UTC
CPU functions, Snapshot restores (beta), and Image builds went down.
- resolved May 19, 2026, 02:04 PM UTC
All services are back.
Internal data stores are fa...
Affected: Function execution (CPU functions)Function execution (GPU functions)Function execution (Web endpoints)Web services (Web API)Storage (Volumes)Sandbox execution (Sandboxes)Image builds (Image builds)
Timeline · 3 updates
- investigating May 08, 2026, 12:22 AM UTC
We are seeing a lot of internal data stores failing right now and are actively working on fixing them.
- resolved May 08, 2026, 02:30 AM UTC
All services except queues have recovered. Queues are currently seeing recovery
- resolved May 08, 2026, 03:50 AM UTC
All data stores are recovered
Container Scheduling and RP...
Affected: Function execution (CPU functions)Function execution (GPU functions)Function execution (Web endpoints)Sandbox execution (Sandboxes)Image builds (Image builds)
Timeline · 2 updates
- investigating May 07, 2026, 11:50 PM UTC
We're experiencing container scheduling failures and some RPC errors due to a storage failure. We're working on mitigating now.
- resolved May 08, 2026, 12:03 AM UTC
We've recovered
H200 Scheduling is degraded
Affected: Function execution (GPU functions)
Timeline · 2 updates
- investigating May 07, 2026, 08:04 PM UTC
Due to an H200 capacity shortage, we are seeing degraded H200 scheduling latency
- resolved May 07, 2026, 10:32 PM UTC
We are still actively working on securing more H200 capacity. We recommend you use RTX6000s or H100s in the meantime. Thank you for your patience!