bench: add qdrant comparison to vector benchmark - #145
Merged
Conversation
adds qdrant as an optional fourth comparison target in the vector similarity benchmark suite. opt-in via --qdrant flag to avoid requiring docker for the default run. - qdrant python client class using official SDK - docker container lifecycle (startup, health check, teardown) - results parsing, display table, and CSV output columns - pip dependency management for qdrant-client
combines qdrant comparison (this branch) with ember-grpc benchmark (from main). both systems now available in choices, flags, and CSV output.
- use query_points instead of deprecated search method - use HnswConfigDiff model instead of raw dict - handle --ember-only --qdrant displaying both columns
kacy
added a commit
that referenced
this pull request
Feb 19, 2026
* bench: add qdrant comparison to vector benchmark adds qdrant as an optional fourth comparison target in the vector similarity benchmark suite. opt-in via --qdrant flag to avoid requiring docker for the default run. - qdrant python client class using official SDK - docker container lifecycle (startup, health check, teardown) - results parsing, display table, and CSV output columns - pip dependency management for qdrant-client * fix qdrant client for current SDK and display table - use query_points instead of deprecated search method - use HnswConfigDiff model instead of raw dict - handle --ember-only --qdrant displaying both columns
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
summary
--qdrantflag so existing usage is unchangedqdrant/qdrantdocker image andqdrant-clientpython SDKwhat was tested
cargo fmtandcargo clippypass cleanlocal results (100k vectors, 128-dim, cosine, k=10)
ember query throughput is 5.3x faster than qdrant and 3.4x faster than pgvector at this scale. qdrant leads on insert throughput thanks to native batch upsert.
chromadb skipped — pydantic v1 is broken on python 3.14.
design considerations
--qdrantflag) rather than always-on to keep docker optional for the default run--ember-only --qdrantworks as a 2-system comparison (no chromadb/pgvector needed)query_pointsAPI (current qdrant SDK;searchwas removed)