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 v2.1.293,
ImprovementsSection of the release
Unclear It is not clear what condition decides when Claude Haiku 4.5 is used.
What
Claude Code uses a small, fast model for some side tasks. When you connect directly to Anthropic's API, it now chooses Claude Haiku 4.5 for that role if a certain condition holds and the model is not excluded. Otherwise it uses the usual default small model.
Why
This could change which small model handles side tasks.
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 what condition decides when Claude Haiku 4.5 is used.
