docs(models): document reasoning_effort levels across the chat catalogue - #91
Merged
Merged
Conversation
New 'Controlling reasoning' section (EN/ES) stating the per-model contract, verified against production on 2026-09-17: glm5.3 family takes low/medium/high/max; qwen3.6 and gemma4 take none/minimal (no reasoning phase) and low/medium/high/max, which cap the reasoning budget at 2,048/8,192/16,384/32,768 tokens; deepseek-v4-flash is adaptive; qwen3.8-flash and mimo-v2.5 accept the parameter and manage their own depth. Card capability lines updated to match, and the API spec reasoning_effort parameter now describes exactly this (it previously claimed deepseek-only with low/medium/high).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds the 'Controlling reasoning' section (EN/ES) to docs/models with the per-model reasoning_effort contract (verified against prod via api.nan.builders on 2026-09-17), updates the model-card capability lines, and rewrites the OpenAPI reasoning_effort description (previously documented as deepseek-only with low/medium/high). 1302 site tests green.