π¦ New version release#220
Merged
Merged
Conversation
29a1f82 to
d204451
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will be updated.
Releases
@smooai/smooth-operator@1.22.10
Patch Changes
22b193e: Fix
eventual_responsestill shipping an empty reply (blankresponseParts+ emptysuggestedNextActions) on gpt-oss-120b via the LiteLLM/groq gateway, which 1.22.1 did not cover.Confirmed empirically against the real SSE parser: this gateway/model emits the WHOLE answer on the reasoning channel (
delta.reasoning_content) withdelta.contentnever populated. The engine accumulates reasoning into a separate buffer and drops it fromresponse.content, so BOTHlast_assistant_content()and the 1.22.1streamed_reply(content tokens) come back empty β even though the answer streams to the client asstream_reasoningand persists. The "streamed tokens" observed in prod werestream_reasoningframes (protocol-identical tostream_token), not content.rust/smooth-operator-server/src/runner.rs: accumulate the turn's reasoning stream and use it as a LAST-RESORT fallback for the final reply β afterlast_assistant_content()andstreamed_reply, only when no answer content exists anywhere. A normal reasoning model always populatescontent, so it never surfaces its thinking as the answer; this rung fires solely for the degenerate answer-in-reasoning case where the alternative is an empty response. The suggested-replies trailer is preserved through the fallback so suggestions are recovered.Adds
tests/gateway_wire_empty_reply.rs, a regression that drives the realLlmClientagainst a local mock speaking the gateway SSE wire format (answer-in-content and answer-in-reasoning shapes) β it fails if the reply goes empty again.@smooai/smooth-operator-web-chat-example@0.0.3
Patch Changes