chicco4life

Thanks for sharing this.

I've been focusing on deep research related tasks for biotch and life science applications. The problem with this sort of task is that we need subagents to reason through multiple (potentially 100s or more) chains of knowledge/concept/evidence, so the token usage really explodes as complexity of the task and data expands. A typical task can cost me nearly a $1k overnight...

I've been testing out GLM5.3, but now I'm really tempted to try to Deepseek 4.1 flash too. Any chance you've benchmarked / compared the two?

Agreed.

I've been running automated research tasks for life sciences companies, and the speed in which tokens are burnt is scary. Especially when you get into a complex knowledge space and require a subwgent to reason through each possibility, token usage grows quadratically not linearly as complexity increases...