6 ms·

I believe this is to improve performance by shortening the context window for long thinking processes. I don't think this is referring to real-time summarizing

by comova 1y ago

I believe this is to improve performance by shortening the context window for long thinking processes. I don't think this is referring to real-time summarizing for the users' sake.

usaar333 1y ago
When you do a chat are reasoning traces for prior model outputs in the LLM context?
- int_19h 1y ago
  No, they are normally stripped out.
j_maffe 1y ago
> I don't think this is referring to real-time summarizing for the users' sake. That's exactly what it's referring to.