Repository navigation
Send reasoning_effort "none" for Kimi K3 thinking Off on Posit AI Pass - #192
Merged
Merged
Conversation
Kimi K3 offers "off" in its capability entry, but positAiThinkingRequestFields only maps "off" to reasoning_effort "none" for DeepSeek V4.1 Flash and Kimi K3.1 Pro. For Kimi K3 the request carried no reasoning field, so Baseten's default effort applied and the model reasoned anyway. Fixes posit-dev#191 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #191.
Kimi K3 offers "off" in its capability entry, but
positAiThinkingRequestFieldsonly sendsreasoning_effort: "none"for DeepSeek V4.1 Flash and Kimi K3.1 Pro. For Kimi K3 the request went out with no reasoning field, so Baseten's default effort applied and the model reasoned anyway. This adds Kimi K3 to the same exception, in the form of #176, plus a repro test next to the DeepSeek one.Checks run on
mainatf456292:expected undefined to deeply equal { reasoning_effort: 'none' }) and passes with it.ai-provider-bridge: vitest, 91 files, 1,112 tests passed;tsc --noEmit; oxlint.oxfmt --checkon the whole repo, and the shape guard (typechecks/tsconfig.json).The measurements (cost and latency on Posit AI Pass with and without
"none") and a capability-flag alternative for future models are in #191. The flag touches the chat params the host passes, so it is not in this PR.🤖 Generated with Claude Code