{"version":"2.1.300","anchor":"new-internal-model-capability-field-supportsthinkingoff","canonical_anchor":"new-internal-model-capability-field-supportsthinkingoff","heading":"Model entries now say whether thinking can be turned off","tier":"internal","area":"Thinking","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.300\/e\/new-internal-model-capability-field-supportsthinkingoff","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.300","markdown":"### Model entries now say whether thinking can be turned off\n\nEach model entry gains a `supportsThinkingOff` field telling programs whether that model allows thinking to be switched off\n\n**What**\n\nThinking is the step where the model reasons before answering. Each model in Claude Code's model list now carries a `supportsThinkingOff` value saying whether thinking can be turned off for that model. The field is optional and marked internal, described as \"Whether thinking can be turned off for this model\".\n\n**Why**\n\nPrograms and interfaces built on Claude Code can use this to tell which models allow thinking to be switched off, for example to disable a thinking toggle for models that do not.\n\n- Area: Thinking\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 3\/5\n- Scope: individual\n- Heads-up: no"}