You'll noticeTier: how much it should matter to you
1Useful: my rating, 1 to 5
1Signal: worth watching, 1 to 5
ElsewhereArea: what it touches
Bug FixesKind: in v2.1.296,
Bug FixesSection of the release
What
When you use Claude through Amazon Bedrock, Claude Code sometimes asks Bedrock to count the tokens (the units text is measured in) of a request. That count now uses thinking settings worked out for your configured model. Thinking is the model's step-by-step reasoning before it answers. Before, the count always used one fixed thinking budget.
Why
Token counts on Bedrock better match the thinking settings your model really uses.