Explore extended thinking mode for Duo Code Review

Everyone can contribute. Help move this issue forward while earning points, leveling up and collecting rewards.

• Label this issue
• Close this issue

Claude 3.7 has a capability to respond with an extended thinking mode enabled. It seems that it could provide more comprehensive code analysis. It would be slightly slower and would use more credit so that should also be considered when making this change.

Here is one of the experiment that showed some improvements and the difference in timing and token usage can be seen. About 20% slower and 10% more token usage in this case.

https://smith.langchain.com/o/477de7ad-583e-47b6-a1c4-c4a0300e7aca/datasets/607673ec-f1a2-496a-a7b8-2f3adf09b212/compare?selectedSessions=5d506e91-a5e6-4c6f-866f-59f1c7729372%2C4d91380a-a2ef-40a4-90ed-1b4a6e702f96%2C856116be-25b2-4684-b18a-69c6da8c9936&baseline=5d506e91-a5e6-4c6f-866f-59f1c7729372&compare-experiment-tab=0&activeSession=5d506e91-a5e6-4c6f-866f-59f1c7729372

See for more detail https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking