fix(vllm): pool embedding worker output via PoolingParams(task="embed") (#10248) [cherry-pick → release/1.3.0] - #11555
Merged
nv-nmailhot merged 1 commit intoJul 11, 2026
Conversation
…") (#10248) Signed-off-by: Tzu-Ling <tzulingk@nvidia.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Signed-off-by: Tzu-Ling <tzulingk@nvidia.com>
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Overview:
Cherry-pick of #10248 into
release/1.3.0.Details:
Backports
fix(vllm): pool embedding worker output via PoolingParams(task="embed")(#10248). The vLLM embedding worker now requests the pooled sentence embedding viaPoolingParams(task="embed")instead of relying on vLLM's default (per-token) encode output, and forwards the OpenAIdimensions(Matryoshka) parameter to vLLM's pooler — which truncates + re-normalizes and validates model support — rather than slicing the vector in Python. Adds unit tests, anagg_embed.shlaunch example, and a serve-test tweak.Cleanly cherry-picked (no conflicts) from merge commit
0b973594e2.Where should the reviewer start?
components/src/dynamo/vllm/handlers.py—EmbeddingWorkerHandler.generatenow buildsPoolingParams(task="embed", dimensions=...)and drops the post-hoc Pythonembedding[:dimensions]truncation.components/src/dynamo/vllm/tests/test_vllm_worker_handler.py—test_dimensions_forwarded_to_pooling_params,test_no_dimensions_omits_pooling_dimensions,test_oversized_dimensions_raises.🤖 Generated with Claude Code