sacha
Founder at Edgee
- Joined
- Nov 18, 2026
Comments
Switching LLMs mid-conversation is a ski-rental problem - Edgee Blog
Great comment, thanks @Primoh. I agree with you about choosing one model per subagent; it’s indeed the simplest and most effective approach. Changing models mid-task through the main conversation poses a very significant risk of breaking the cache. However, our research has shown that with a rigorous approach to classifying tasks and determining the moment at which we can switch models, this can become very economical. Particularly when a developer uses the same session to carry out several tasks one after another.
Switching LLMs mid-conversation is a ski-rental problem - Edgee Blog
hey @nate, Khaled PhD is a real human :) Feel free ton contact him through Linkedin to have a discussion about implementation details
[Weekly] What are you building this week?
I’m building a judge model, a coding task classifier. Super fun. Feel free to give me ideas to get coding session dataset ;)
Switching LLMs mid-conversation is a ski-rental problem - Edgee Blog
Happy to share the latest work of our Head of Research about how and when to switch models in a coding task. No that simple, and fun to see how scientist can handle this problem.