Repository navigation
docs(models): the output floor covers max_output_tokens too - #199
Merged
Merged
Conversation
The deepseek-v4-flash card said "a max_tokens below 16384 is raised to 16384", which described only the Chat Completions fields. The floor now covers the Responses API's max_output_tokens as well, so one sentence covers both routes and states the other half of the rule: a budget above 16384 is never lowered.
Completes the previous commit, which only touched the English card:
* the Spanish card (es/models.mdx) said the same thing about
max_tokens only, so a reader of either locale now sees the same
rule, including "a budget above 16384 is never lowered";
* openapiSpec.test.ts and docsReasoning.test.ts assert the card text
verbatim -- by design -- so they had to move with it. They now also
pin the Responses-API half and the never-lowered half in both
locales, which is what would have caught a half-updated card.
Full suite: 2496 tests / 87 files green. npm run build green.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
The
deepseek-v4-flashcard documented the output-budget floor as "amax_tokensbelow 16384 is raised to 16384", which named only the Chat Completions fields.The floor now covers the Responses API's
max_output_tokensas well (see helmcode/nan-devops#893), so the sentence covers both routes and states the other half of the rule explicitly: a budget above 16384 is never lowered.Why
The same model answered with a different output budget depending on the route, and the public doc described neither the parity nor the "never lowered" half. A client reading only this page could not tell what their
max_output_tokenswould become.Verification
npm run buildpasses in the worktree.