feat(perf): declare enable-thinking controls without changing legacy requests - #19
Merged
Merged
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #11.
Adds an explicit, closed
request.thinking_controldeclaration:{"kind":"vllm-enable-thinking-v1","enabled":false}undervllm-fixed-v1.It sends only
chat_template_kwargs.enable_thinking. Existingrequest.thinkingcontinues sending onlychat_template_kwargs.thinking; old normalized workload hashes and regenerated request bytes remain unchanged. Two non-null declarations are rejected, including agreeing values. Unknown kinds/fields and provider-specific controls underportable-chat-v1are rejected before dispatch.Proposal reconciliation: portable-chat-v1 remains provider-neutral; this provides the requested recorded control via the explicit vLLM profile rather than weakening the portable contract. No generic extra_body or automatic model-name mapping. Declared thinking-off is not proof of provider compliance; the collector retains observed answer/reasoning evidence without inventing success.
Verified: independent SIX and HELP source reviews; exact-head fmt, Clippy -D warnings, full workspace tests and release build; real-CLI loopback request/identity/conflict regressions. Integrated smoke exercised the new field and old saved receipts. No inference-server call or new live template qualification occurred.