Discovered this after couple of hours of being confused.
In normal chats, Claude desktop doesnt even display API Error: 529. It only does so in Cowork tasks. The regular chat just does an infinite "Clauding" animation.
Contrast with Claude Government with all green 100% uptime. So either no one is using it or the people within Anthropic running that one have different priorities.
Same here, our Claude API calls started throwing 529s this morning. Added a simple retry-with-backoff and it recovered fine once the incident cleared, but definitely caught us off guard.
Discovered this after couple of hours of being confused. In normal chats, Claude desktop doesnt even display API Error: 529. It only does so in Cowork tasks. The regular chat just does an infinite "Clauding" animation.
Subscribed to Cursor again. UI now looks the same. Literally zero switching costs. Crazy how commoditized this is.
Again, again and again.
Contrast with Claude Government with all green 100% uptime. So either no one is using it or the people within Anthropic running that one have different priorities.
Hopefully they are fixing fable burning through entire sessions in minutes
Same here, our Claude API calls started throwing 529s this morning. Added a simple retry-with-backoff and it recovered fine once the incident cleared, but definitely caught us off guard.
It has been like that for the last 2 hours.
Claude is a one 9 service.
Seems to be back up now, I've had a couple messages go through successfully in the past ~5min.
Claude late to work again? And especially on a Monday it is now down? Oh dear.
Why don’t you give Claude a break, non-stop asking too many questions at it and putting it in hundreds of loops to cause it to crash out.
Clearly Claude can’t handle that.
> At this time, we have seen errors stabilize
"Stabilized errors". Hard to tell what Anthropic's difficulty is here. Service or English.
[flagged]