Hello, a lot of times when using opus 4.6, it thinks very very long (multiple minutes) even if it is a simple problem. When I expand the thinking accordion to see what it is doing, it is always stuck on stuff like this:
My question is, is this thinking even useful? Does it actually do the things it is thinking about? Or is this a bug
Thank you in advance
Got the exact same issue and it is driving me crazy! Considering switching to Claude Code.
I’ve seen this happen only once I think - rest of the time it works fine. Weird.
Share your request ID with your message so that they can investigate.
Same issue here too
I’ve seen this happen sometimes too. Opus tends to overthink simple prompts, so it can get stuck in extended reasoning even when it’s not really necessary.
The “thinking” you see isn’t always a literal process the model is executing step by step it’s more like a reasoning trace. Sometimes it’s useful, but other times the model just spends too long exploring possibilities.
Usually the easiest fixes are:
cancel and resend the prompt
keep the prompt a bit more direct/simple
or switch models and try again
If it happens a lot with simple tasks, it might also be a temporary backend or streaming issue
Hey, thanks for the report. This is a known issue with Opus 4.6. The model sometimes gets stuck in a long reasoning loop, even on simple tasks.
On your question, the thinking trace is the model’s real reasoning process. But when it loops for minutes on a simple task, it usually doesn’t help much. The team is aware of the issue.
For now, you can try:
Can you share the Request ID from the chat where this happened? Click the three dots in the top-right of the chat, then Copy Request ID. This will help us figure out what’s going on in the backend.
