Add colbert-pooling sample app: ColBERT with hierarchical token pooling#1889
Draft
thomasht86 wants to merge 1 commit intomasterfrom
Draft
Add colbert-pooling sample app: ColBERT with hierarchical token pooling#1889thomasht86 wants to merge 1 commit intomasterfrom
thomasht86 wants to merge 1 commit intomasterfrom
Conversation
PoolingColBertEmbedder runs ColBERT-Zero ONNX model inside Vespa and applies Ward's agglomerative clustering to reduce per-document token count. Binary pooled embeddings are HNSW-indexed in memory for fast ANN retrieval; full bfloat16 embeddings stay paged on disk for exact second-phase reranking. ColBertSearcher handles query embedding, binarisation, and nearestNeighbor query construction so users just send text. Verified against pylate (model authors' reference): cos_sim > 0.999 per token vs ONNX INT8 reference, 100% bit-exact binarisation. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
thomasht86
temporarily deployed
to
Vespa Cloud CD
GitHub Actions
Inactive
bratseth
reviewed
Apr 13, 2026
Member
There was a problem hiding this comment.
This isn't needed since the inputs are specified in the rank profile.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
PoolingColBertEmbedder runs ColBERT-Zero ONNX model inside Vespa and applies Ward's agglomerative clustering to reduce per-document token count. Binary pooled embeddings are HNSW-indexed in memory for fast ANN retrieval; full bfloat16 embeddings stay paged on disk for exact second-phase reranking.
ColBertSearcher handles query embedding, binarisation, and nearestNeighbor query construction so users just send text.
Verified against pylate (model authors' reference): cos_sim > 0.999 per token vs ONNX INT8 reference, 100% bit-exact binarisation.
I confirm that this contribution is made under the terms of the license found in the root directory of this repository's source tree and that I have the authority necessary to make this contribution on behalf of its copyright owner.