Daniel San(@dani_avila7):Don’t switch models mid-session Keep one model as the executor and use stronger models as advisors on demand Or delegate specific tasks to subagents running different models

2026.08.25 11:45

Don’t switch models mid-session Keep one model as the executor and use stronger models as advisors on demand Or delegate specific tasks to subagents running different models

Teknium 🪽@Teknium

2026.08.25 06:48

Just FYI. If you are switching models in a session all the time - you are doing it wrong. Every time you switch, your entire prompt cache is invalidated on the new model you switch to, and you have to repay the full input tokens price for all of it. Stop doing this unless those models are free. This is not a hermes thing - this is a fundamentals of inference thing.