-
Fixed manual budget_tokens thinking budget on every request
Setting a static thinking.budget_tokens value on every API request to control how much a model reasons before answering -- the original mechanism for extended thinking on Claude 4-era models.
Superseded by
adaptive-thinking-effort-controlWhy this verdict?
Anthropic has deprecated this on Claude 4.6 models and removed it entirely on Claude 4.7+, where requests using type: 'enabled' now return a 400 error, in favor of adaptive thinking plus the effort parameter. Still required on models that only support extended-thinking mode (Opus 4.5, Haiku 4.5, Sonnet 4.5 and earlier), so not dead everywhere, but no longer the recommended path for new work.
-
Adaptive thinking + effort control over a fixed manual thinking budget
On current reasoning-enabled Claude models, let the model decide whether and how much to 'think' per request (thinking: {type: 'adaptive'}) and steer depth with a coarse effort setting (low/medium/high), instead of hand-setting a fixed budget_tokens value on every call.
Why this verdict?
Anthropic's own docs mark the manual budget_tokens mechanism as deprecated on Claude 4.6 models and rejected outright (400 error) on Claude 4.7 and later -- adaptive thinking is the only mode going forward wherever it's available. Adaptive mode can also skip thinking entirely on easy inputs at lower effort levels, which a fixed budget can't do.