You'll noticeTier: how much it should matter to you
2Useful: my rating, 1 to 5
3Signal: worth watching, 1 to 5
ModelsArea: what it touches
ImprovementsKind: in v2.1.323,
ImprovementsSection of the release
Unclear It is not clear whether the actual default limits are different for any model.
What
Output tokens are the units that measure how long a reply from the model can be. Claude Code now works out the default limit on output tokens, and the default effort level, from a value set for each model. The older way of calculating these defaults has been removed.
Why
The default maximum reply length could be different for some models.
How sure we are
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubtIt is not clear whether the actual default limits are different for any model.
