Claude may skip ‘thinking’ even when it is turned on
4.7 and 4.8 appear to stop using the ‘thinking’ feature consistently in some conversations. At first, the model uses deeper analysis two or three times.
After that, it starts replying almost instantly, even though thinking is turned on and set to . The usage pattern has not changed, and this behavior had not appeared for months before now.
Key points
- 4.7 and 4.8 may stop using ‘thinking’ after the start of a conversation.
- The first few replies can show deeper analysis, then later replies arrive instantly.
- The issue appears even when thinking is enabled and set to .
- The reported did not change from earlier use.
- For important development work, fast answers should be checked more carefully.