{"version":"2.1.295","anchor":"model-max-output-token-lookup-now-takes-the-modelcontext-ar","canonical_anchor":"model-max-output-token-lookup-now-takes-the-modelcontext-ar","heading":"Maximum response length lookup takes more information into account","tier":"internal","area":"Elsewhere","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.295\/e\/model-max-output-token-lookup-now-takes-the-modelcontext-ar","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.295","markdown":"### Maximum response length lookup takes more information into account\n\nThe fallback that sets a model's maximum response length now receives more information, so the limit can depend on more than the model\n\n**What**\n\nEvery model has a cap on how long a single reply can be, measured in tokens (the small chunks of text a model reads and writes). Claude Code has changed how it looks that cap up. The fallback step, used when no specific value is found, now receives the same information as the first step, so the limit it picks can depend on more than the model alone.\n\n**Why**\n\nThe maximum reply length could come out differently for some models.\n\n- Area: Elsewhere\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}