{"version":"2.1.248","anchor":"reasoning-effort-levels-and-their-aliases","canonical_anchor":"reasoning-effort-levels-and-their-aliases","heading":"Reasoning effort levels and their aliases","tier":"use","area":"Model Settings","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248\/e\/reasoning-effort-levels-and-their-aliases","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.248","markdown":"### Reasoning effort levels and their aliases\n\nReasoning effort accepts low through max, with med and ultracode as aliases.\n\n**What**\n\nReasoning effort accepts `low`, `medium`, `high`, `xhigh` and `max`, with `med` accepted as an alias for medium and `ultracode` as an alias for xhigh.\n\n**Details**\n\n- A related state object gained `unusableWritingOverrideTold`, defaulting to false, tracking whether a one-time notice has been shown.\n\n**Evidence**\n\n`[\"low\", \"medium\", \"high\", \"xhigh\", \"max\"]`\n\n- Area: Model Settings\n- Names: `low`, `medium`, `high`, `xhigh`, `max`\n- Tier: Use it now\n- Useful: 4\/5\n- Signal: 4\/5"}