DX-596-together-chat-completions: sync with mintlify-docs#1159#39
Merged
zainhas merged 1 commit intoJul 10, 2026
Merged
Conversation
Sync clarifications from togethercomputer/mintlify-docs#1159: - Note that the `reasoning` field is symmetric on input and output for preserved thinking / multi-turn tool calling; `reasoning_content` is still accepted on input for backward compatibility. - Document that reasoning token and cached-token counts on `usage` vary in location by model (nested `*_details` on reasoning models, flat `cached_tokens` on some non-reasoning models) and add a defensive-read pattern so clients don't silently see 0.
blainekasten
approved these changes
Jul 10, 2026
zainhas
deleted the
docs-sync/together-chat-completions/mintlify-docs-pr-1159
branch
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Syncs the
together-chat-completionsskill with mintlify-docs #1159 — Clarify reasoning field + usage-object shape, fix DeepSeek V4 Pro pricing.Triggering docs changes
docs/inference/chat/reasoning.mdxdocs/inference/openai-compatibility.mdxdocs/deepseek-v4-quickstart.mdx(pricing-only; not tracked in skill'sreferences/models.md, no sync needed)Skill changes
references/reasoning-models.mdreasoningis symmetric between input and output. When you pass a prior assistant turn back to the API (preserved thinking, multi-turn tool calling), use thereasoningkey on the assistant message.reasoning_contentis still accepted on input for backward compatibility.reasoning_tokensandcached_tokenslive in different places depending on the model — reasoning models nest them underusage.completion_tokens_details.reasoning_tokensandusage.prompt_tokens_details.cached_tokens; some non-reasoning models returncached_tokensflat at the top ofusage. Includes a Python + TypeScript defensive-read pattern so clients don't silently see0.SKILL.mdusagetoken counts).The DeepSeek V4 Pro pricing change ($1.74 / $3.48 / $0.20 per 1M in/out/cached tokens) is not carried over because
references/models.mdhas no pricing columns and no other skill file references those numbers.Generated by the Sync Skills Cursor Automation. Please review before merging.