Join Nostr
2026-08-11 13:15:26 UTC
in reply to

Silberengel on Nostr: Yes, but it sometimes glosses over failures. I have to point it out and then it's ...

Yes, but it sometimes glosses over failures. I have to point it out and then it's like, Oh, right, that isn't how we originally described it.

Or I give it 15 points and it does 1-4, 6-9, 10-15. Have to have it go back and review a few times, sometimes switching models, to get full and complete and correct coverage.