Nothing to try yetTier: how much it should matter to you
1Useful: my rating, 1 to 5
3Signal: worth watching, 1 to 5
Model ConfigArea: what it touches
In DevelopmentKind: in v2.1.234,
In DevelopmentSection of the release
A model-config prefetch step runs at startup but its guard is unconditionally false, so nothing is fetched.
What
Startup now calls a step that would look up the chosen model through the API and cache its token limits before the first turn. Its guard returns false unconditionally in this build, so the call returns immediately and nothing is fetched.
Details
When enabled it would log the resolved max_input_tokens= and max_tokens= values under a [modelConfig] prefix.
None of this code exists in v2.1.233.
Evidence
[modelConfig]
Strings lifted out of the shipped bundle, so the claim above can be checked against them.
