{"version":"2.1.275","anchor":"modelcomplete-maxtokens-is-now-validated-against-the-actu","canonical_anchor":"modelcomplete-maxtokens-is-now-validated-against-the-actu","heading":"$.model.complete maxTokens is now validated against the actual model's output ceiling, not a fixed global cap","tier":"use","area":"Plugin API","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.275\/e\/modelcomplete-maxtokens-is-now-validated-against-the-actu","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.275","markdown":"### $.model.complete maxTokens is now validated against the actual model's output ceiling, not a fixed global cap\n\n$.model.complete's maxTokens limit now matches each model's real output ceiling instead of one fixed number\n\n**What**\n\nThe hook API's `$.model.complete` previously rejected any `maxTokens` value above a single hardcoded limit of 8192, regardless of which model was being used. It now looks up the actual output ceiling for the specific model being called (still capped globally at a larger number) and rejects requests that exceed it with 'maxTokens ${s} is past what ${w} can produce in one reply (${O})'.\n\n**Why**\n\nModels that can produce longer replies than 8192 tokens are no longer artificially capped, while models with smaller limits are still protected from requesting more than they can actually return.\n\n- Area: Plugin API\n- Names: `maxTokens`\n- Tier: Use it now\n- Useful: 3\/5\n- Signal: 1\/5"}