Theo-brainUnofficial. Not affiliated with Theo or t3.gg.
Sign inYou

Trim slow tool calls when the model is fast

Technique

Tool calls are the actions an agent takes besides writing, such as git commands, builds or browser snapshots. When a fast model writes in 30 seconds instead of 10 minutes, these slow steps can double the total wait, so it pays to cut them.

Theo's current stance

Theo told his agent to only change code and leave the preview browser alone so updates show up immediately. He even wants a separate system prompt for Ultrafast.

Try it

  1. Notice which agent steps take seconds without changing your code.
  2. Tell the agent to skip them, e.g. 'just change code when I ask, don't drive the preview browser.'

First seen

7:29
when inference goes from 10 minutes to 30 seconds, that 30 seconds of waiting on tool calls goes from a small percentage of your runtime to doubling your runtime. So, I just told it, for example, to not make changes to the preview browser, just change code when I ask, because I don't want it to call tools that take seconds, because I just want to see the updates while we go. So,

I love Ultrafast (it's unusable)

Every mention

7:29
when inference goes from 10 minutes to 30 seconds, that 30 seconds of waiting on tool calls goes from a small percentage of your runtime to doubling your runtime. So, I just told it, for example, to not make changes to the preview browser, just change code when I ask, because I don't want it to call tools that take seconds, because I just want to see the updates while we go. So,

I love Ultrafast (it's unusable)

Theo Daily is unofficial and not affiliated with Theo or t3.gg. Written by Claude Opus 5.5 from each video's transcript.