So I was using Auto mode for research tasks and at some point Cursor decided that it’s going to use claude-4.6-opus-high-thinking for all subagents / skills, etc. As a result a single prompt used about 300M of API tokens (essentially 70% of my monthly Opus quota). [image] I thought by selecting …

Same symptoms? You pick Auto in the model selector and it starts using claude-4.6-opus? Can you share you triggered that?

Unexpected Claude Opus usage in 'Auto' mode - spent 70% of Ultra with a single prompt

Guillermo_Chavez (Guillermo Chavez) February 27, 2026, 12:12am 3

I got the same issue today not sure if is a issue or what

Topic		Replies	Views
Auto is Good Now! Discussions	8	1287	December 21, 2025
Automatic switch to Claude 4.6 Bug Reports auto-mode , anthropic , subagents	1	82	August 1, 2026
Cursor AI Experience: Auto Mode, Model Reliability, and Pricing Transparency Discussions	21	3084	November 17, 2025
Cannot use default model in latest cursor-cli version on grandfathered plan Bug Reports	15	881	March 21, 2026
Documentation inconsistency: Auto mode limit handling needs clarification Feedback	14	2558	August 6, 2025