DuckDB community extension for prefiltered HNSW using ACORN-1
Details
- External ID
- 47512891
- Source
- HN
- Company
- —
- Product
- DuckDB community extension for prefiltered HNSW using ACORN-1
- Website domain
- github.com
- Launched
- March 25, 2026
- Cohort
- —
- Upvotes
- 90
- Upvotes percentile
- 0.897908979089791
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Hey folks! As someone doing hybrid search daily and wishing I could have a pgvector-like experience but with actual prefiltered approximate nearest neighbours, I decided to just take a punt on implementing ACORN on a fork of the DuckDB VSS extension. I had to make some changes to (vendored) usearch that I'm thinking of submitting upstream. But this does the business. Approximate nearest neighbours with WHERE prefiltering.Edit: Just to clarify, this has been accepted into the community extensions repo. So you can use it like:```INSTALL hnsw_acorn FROM community;LOAD hnsw_acorn;```
Enrichment
- Theme
- self-hosted infrastructure and security tools
- Vertical
- —
- Function
- Data infrastructure
- Audience
- Developer
- AI stance
- AI feature
- Project type
- Hobby / open-source project
- Normalized one-liner
- duckdb vector search extension
- Manually corrected
- False
Could you build this?
No Implementing the ACORN algorithm for prefiltered vector search inside a C++ DuckDB/usearch extension requires deep low-level algorithm implementation and database engine internals knowledge.
What it would actually take: This requires modifying low-level C++ codebases (vendored usearch and DuckDB VSS) to implement graph-based approximate nearest neighbor search (ACORN-1) with custom predicate traversal logic during HNSW construction and search. The engineer must understand vector index graph invariants, cache-friendly SIMD distance metrics, and DuckDB's execution engine memory model and extension APIs. This necessitates specialized expertise in database systems internals and advanced vector search algorithmic research.
Discussion
5 comments analyzed.
Competitors mentioned: DuckDB with Lance extension, RaBitQ
Concerns raised: Performance with strict filters affecting speed vs accuracy tradeoff
Feature requests: Efficient binary vectors support, Upstream integration
Competitors
Other products that read as similar to this one — 21 launches clear the similarity bar, closest 8 shown.
Attention rank: #2 of 22 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 106 days after the earliest competitor.
- Ducktape · hn · 2025-12-16 · 12 upvotes · similarity 0.41
- Connect DuckDB to any database that has an ADBC driver · hn · 2026-07-08 · 7 upvotes · similarity 0.39
- Glintlog · hn · 2026-02-04 · 10 upvotes · similarity 0.39
- Dux, distributed DuckDB-backed dataframes on the Beam · hn · 2026-03-31 · 7 upvotes · similarity 0.37
- Snowflake Emulator · hn · 2026-01-03 · 6 upvotes · similarity 0.34
- acorn · hn · 2026-02-24 · 5 upvotes · similarity 0.34
- Ducklang: Achieving 100x more requests per second than NextJS · hn · 2026-01-01 · 9 upvotes · similarity 0.33
- Local Privacy Firewall-blocks PII and secrets before ChatGPT sees them · hn · 2025-12-09 · 111 upvotes · similarity 0.33
Other launches for this product
- No other launches for this product.