MChatzakis/DARTH

[SIGMOD 2026] DARTH: Declarative Recall Through Early Termination for Approximate Nearest Neighbor Search.

37
/ 100
Emerging

When searching through large datasets for similar items, like finding similar images or text, you often have to choose between getting perfect results slowly or faster results that might miss some matches. This tool helps data scientists and machine learning engineers get fast results that consistently meet a pre-defined quality standard, avoiding manual tuning and ensuring important matches aren't missed. It takes your existing similarity search index and desired result quality as input, and outputs significantly faster searches.

No commits in the last 6 months.

Use this if you need to perform approximate nearest neighbor searches on large datasets and want to guarantee a certain level of result quality without sacrificing search speed or manually tuning complex parameters for different queries.

Not ideal if your application requires perfectly exact nearest neighbor search results every time, or if your datasets are small enough that the performance gains from approximation are not significant.

similarity-search information-retrieval large-scale-data machine-learning-engineering data-science
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 6 / 25
Maturity 15 / 25
Community 14 / 25

How are scores calculated?

Stars

23

Forks

4

Language

C++

License

MIT

Last pushed

Aug 19, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/vector-db/MChatzakis/DARTH"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.