3 ms·
Seems very promising. I've used Claude a few times and the responses seemed significantly better than GPT-4.
by wyem 3y ago
Seems very promising. I've used Claude a few times and the responses seemed significantly better than GPT-4.
- cl42 3y agoI don't have GPT-4 API access so can't comment on that yet. The code quality via chat has been comparable. One of my goals in the coming weeks (when I get access) is to formally run tests across the models.