Actually, in the past few days o3 has proven fairly unreliable for me. I've gone back to o1-pro. But when I wrote the above it was reasonably reliable.
o3 with a pdf or in deep research mode is excellent. Especially if you’re disciplined about staying to what’s research. But really, it’s excellent, better than benchmarks indicate, I’d say.
Comments
ChatGPT o3 understands the latest literature and isn't going to hallucinate weird details or make incorrect analogies or math?
I'd worry about learning the wrong things.
Actually, in the past few days o3 has proven fairly unreliable for me. I've gone back to o1-pro. But when I wrote the above it was reasonably reliable.
I disagree. It's all about rephrasing information that is in the paper. Possinly a few other papers too.
o3 with a pdf or in deep research mode is excellent. Especially if you’re disciplined about staying to what’s research. But really, it’s excellent, better than benchmarks indicate, I’d say.