Google is winning on every AI front

In my experience Claude 3.7 is far superior for coding than Gemini 2.5. I tried it in Cursor and I wanted it to work, as a recent ex-Googler. I repeatedly found it inferior. I think it’s still behind Claud 3.5 for coding.

It would decide arbitrarily not to finish tasks and suggest that I do them. It made simple errors and failed to catch them.

Your issue is because:

1- the cursor agent doesn’t work with gemini. Some times the diff edit even doesn’t work.

2- Cursor does semantic search to lower the token they sent to models.

The big advantage for Gemini is the context window, use it with aider, clien or roo code.