{"version":"2.1.280","anchor":"thinking-blocks-now-kept-across-arbitrary-model-changes-not","canonical_anchor":"thinking-blocks-now-kept-across-arbitrary-model-changes-not","heading":"Thinking blocks now kept across arbitrary model changes, not just upgrades","tier":"soon","area":"Thinking Mode","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.280\/e\/thinking-blocks-now-kept-across-arbitrary-model-changes-not","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.280","markdown":"### Thinking blocks now kept across arbitrary model changes, not just upgrades\n\nClaude Code can now keep extended-thinking content across any model switch, not just upgrades\n\n**Unclear.** Whether this gate is enabled for any given account, including this site's, is not established by the finding.\n\n**What**\n\nWhen you switch the model Claude Code is using mid-session, the app has to decide what to do with 'extended thinking' blocks (the model's visible step-by-step reasoning) produced by the previous model. Previously this was a simple yes\/no choice, and thinking blocks were only kept when the switch was an upgrade to a newer model. That setting now has three possible values:\n\n- `all` - keep thinking blocks across any model change, upgrade or not\n\n- `upgrade` - keep them only when switching to a newer model, the previous behaviour\n\n- `none` - drop them\n\nA new telemetry event, `api_keep_thinking_on_model_change`, is recorded when the `all` behaviour applies.\n\n**Why**\n\nThis gives more flexibility around preserving a model's prior reasoning when you change models, rather than only on upgrades. It's controlled by a gate (an internal switch Anthropic uses to turn features on for some accounts), so which behaviour applies isn't something you can tell from this release alone.\n\n- Flag `tengu_rustling_pixel`: Not enough to say (read for one account on one subscription tier against v2.1.280; this account: no value returned, anonymous baseline: no value returned, compiled default: not a boolean we can read) These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.\n- Area: Thinking Mode\n- Tier: Nothing to try yet\n- Useful: 3\/5\n- Signal: 3\/5\n- Present in the build but not switched on"}