Open Issues Need Help
View All on GitHub Add a Markdown/CSV export of the ranked-config table about 2 hours ago
enhancement good first issue
Benchmark RAG retrieval configs on your own corpus and get deterministic, stage-level attribution for every failed query.
Python
#ai#benchmark#bm25#embeddings#evaluation#hnsw#hybrid-search#information-retrieval#llm#python#rag#reranking#retrieval-augmented-generation#vector-search
Add an API embedder backend (OpenAI / Cohere) behind an extra about 2 hours ago
enhancement help wanted
Benchmark RAG retrieval configs on your own corpus and get deterministic, stage-level attribution for every failed query.
Python
#ai#benchmark#bm25#embeddings#evaluation#hnsw#hybrid-search#information-retrieval#llm#python#rag#reranking#retrieval-augmented-generation#vector-search
Add a generic CSV → verified-gold importer about 2 hours ago
enhancement help wanted good first issue
Benchmark RAG retrieval configs on your own corpus and get deterministic, stage-level attribution for every failed query.
Python
#ai#benchmark#bm25#embeddings#evaluation#hnsw#hybrid-search#information-retrieval#llm#python#rag#reranking#retrieval-augmented-generation#vector-search
Add an MS MARCO dataset importer about 2 hours ago
enhancement help wanted good first issue
Benchmark RAG retrieval configs on your own corpus and get deterministic, stage-level attribution for every failed query.
Python
#ai#benchmark#bm25#embeddings#evaluation#hnsw#hybrid-search#information-retrieval#llm#python#rag#reranking#retrieval-augmented-generation#vector-search
Add an MMR (diversity) reranker about 2 hours ago
enhancement good first issue
Benchmark RAG retrieval configs on your own corpus and get deterministic, stage-level attribution for every failed query.
Python
#ai#benchmark#bm25#embeddings#evaluation#hnsw#hybrid-search#information-retrieval#llm#python#rag#reranking#retrieval-augmented-generation#vector-search
Add a sliding-window (overlap) chunker about 2 hours ago
enhancement good first issue
Benchmark RAG retrieval configs on your own corpus and get deterministic, stage-level attribution for every failed query.
Python
#ai#benchmark#bm25#embeddings#evaluation#hnsw#hybrid-search#information-retrieval#llm#python#rag#reranking#retrieval-augmented-generation#vector-search