retrieval-benchmark

Here are 7 public repositories matching this topic...

hopit-ai / Moda

MODA: open fashion retrieval benchmark and models by Hopit AI. MODA (203M, open source), MODA Pro Lite (213M, open weights), MODA Pro (hosted). Full-corpus benchmarks vs FashionSigLIP, SigLIP-SO400M and ZooClaw — one harness, losses shown. #1 open model on LookBench.

Updated Aug 28, 2026
Python

eyuansu62 / agent-retrieval-bench

Star

Agentic code retrieval benchmark for coding agents

benchmark software-engineering code-retrieval coding-agents agentic-retrieval retrieval-benchmark

Updated Oct 27, 2026
Python

firecrawl / benchmark-legal-retrieval

Star

benchmark retrieval search-api regulations web-search statutes legal-search firecrawl official-sources retrieval-benchmark legal-retrieval primary-law government-search

Updated Nov 4, 2026
Python

GoparapukethaN / rag-forge

Star

RAG retrieval benchmark runner with JSON reports, Pareto plots, and regression gates for retrieval quality changes.

python nlp benchmark information-retrieval embeddings regression-testing bm25 reranking quality-gates rag llm llmops retrieval-augmented-generation retrieval-evaluation retrieval-benchmark

Updated May 20, 2026
Python

LaimoKarca / fuzzy-travel-recall-benchmark

Star

A reproducible benchmark for incomplete-cue personal mobility recall, comparing BM25, multilingual E5 dense retrieval, and dense-seeded graph retrieval with RRF/PPR diagnostics.

information-retrieval reproducible-research networkx bm25 personalized-pagerank mobility-data dense-retrieval reciprocal-rank-fusion personal-memory multilingual-e5 graph-retrieval retrieval-benchmark multilingual-retrieval lifelog-retrieval

Updated Sep 15, 2026
Python

emadaed / quran-retrieval-benchmark

Star

A 232-question hand-reviewed benchmark for verse-level Quranic retrieval, across six diagnostic buckets (lexical, semantic, multi-hop, cross-lingual, Roman-Urdu, out-of-corpus). Published with the reference pipeline's measured results — including the parts where it fails. CC-BY-4.0.