{"version":"2.1.221","anchor":"output-effort-clamped-when-thinking-is-mechanically-disabled","canonical_anchor":"output-effort-clamped-when-thinking-is-mechanically-disabled","heading":"Output effort clamped when thinking is mechanically disabled","tier":"notice","area":"Models","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.221\/e\/output-effort-clamped-when-thinking-is-mechanically-disabled","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.221","markdown":"### Output effort clamped when thinking is mechanically disabled\n\nInternal helper requests no longer fail from asking for high effort with thinking off.\n\n**What**\n\nInternal one-shot model calls now mark their disabled thinking as mechanical, and a high output effort on such a request is clamped rather than sent and rejected.\n\n**Details**\n\n- Affects web-search assistant calls, hook and classifier prompts, and structured-output helpers.\n- These requests pass `thinkingConfig: { type: \"disabled\", mechanical: !0 }`.\n- When `output_config.effort` would otherwise be high, it is lowered and the clamp is logged, because the API rejects higher effort with thinking disabled.\n\n**Evidence**\n\n`mechanically-disabled thinking, and the API rejects higher effort when thinking is disabled (gh-79798)`\n\n- Area: Models\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 0\/5"}