Investor Nathan Benaich questioned if this coding-focused routing generalizes to other workflows.
@simonw think this applies to non-coding tasks too?
The most interesting Fable tip I've heard so far is to let the model use its own judgement as much as possible
I told it "For all coding tasks use your judgement to decide an appropriate lower power model and run that in a subagent" and it seems to be saving a lot of tokens
Users praise letting AI agents pick their own lower-power submodels as a janky but highly effective trick for cutting token costs, likening it to model merging that somehow just works.
No Digg Deeper questions have been answered for this story yet.
@torchcompiled Letting the model pick its own subagent is the jankiest most effective trick. Same energy as model merging somehow just working