{"data":{"slug":"raudaschl-rag-fusion","name":"rag-fusion","tagline":"multi-query generation + Reciprocal Rank Fusion for retrieval-augmented generation","github_url":"https://github.com/Raudaschl/rag-fusion","owner":"Raudaschl","repo":"rag-fusion","owner_avatar_url":"https://avatars.githubusercontent.com/u/6103284?v=4","primary_language":"Python","stars":952,"forks":115,"topics":["chromadb","information-retrieval","openai","python","rag","rag-fusion","reciprocal-rank-fusion","retrieval-augmented-generation","vector-search"],"archived":false,"github_pushed_at":"2026-04-26T18:45:05+00:00","maintenance_label":"Slowing","stars_delta_30d":6,"url":"https://www.graphcanon.com/tools/raudaschl-rag-fusion","markdown_url":"https://www.graphcanon.com/tools/raudaschl-rag-fusion.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/raudaschl-rag-fusion","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=raudaschl-rag-fusion","description":"RAG-Fusion: multi-query generation + Reciprocal Rank Fusion for better retrieval-augmented generation. Includes evaluation harness with NFCorpus/BEIR.","homepage_url":null,"license":"MIT","open_issues":0,"watchers":14,"ai_summary":"RAG-Fusion uses multi-query generation and Reciprocal Rank Fusion to improve retrieval-augmented generation applications. It provides an evaluation framework using NFCorpus/BEIR.","readme_excerpt":"## Getting Started\n\n1. Install dependencies:\n   ```bash\n   pip install openai chromadb python-dotenv tqdm tabulate rank_bm25\n   ```\n\n2. Set up your OpenAI API key:\n   ```bash\n   cp .env.example .env\n   ```\n   Then edit `.env` and replace `your-key-here` with your actual key.\n\n3. Run the demo:\n   ```bash\n   python main.py\n   ```\n\n4. Run the tests (no API key needed):\n   ```bash\n   python -m pytest test_main.py -v\n   ```","github_created_at":"2023-09-25T21:58:31+00:00","created_at":"2026-07-11T11:34:41.45419+00:00","updated_at":"2026-08-23T06:02:05.523661+00:00","categories":[{"slug":"data-retrieval","name":"Data & Retrieval","url":"https://www.graphcanon.com/categories/data-retrieval","markdown_url":"https://www.graphcanon.com/categories/data-retrieval.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/data-retrieval"},{"slug":"evaluation-observability","name":"Evaluation & Observability","url":"https://www.graphcanon.com/categories/evaluation-observability","markdown_url":"https://www.graphcanon.com/categories/evaluation-observability.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/evaluation-observability"}],"tags":[{"slug":"chromadb","name":"chromadb"},{"slug":"information-retrieval","name":"information-retrieval"},{"slug":"openai","name":"openai"},{"slug":"python","name":"python"},{"slug":"rag-fusion","name":"rag-fusion"},{"slug":"reciprocal-rank-fusion","name":"reciprocal-rank-fusion"},{"slug":"retrieval-augmented-generation","name":"retrieval-augmented-generation"},{"slug":"vector-search","name":"vector-search"}],"trust":{"provenance":{"is_fork":false,"github_id":696503950,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-08-23T06:02:04.771Z","maintenance":{"label":"Slowing","score":36,"methodology":"github_public_v1","releases_90d":0,"days_since_push":118,"last_release_at":null,"stars_delta_30d":6,"open_issues_delta_30d":0},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:34:42.715Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-08-23T06:02:05.228Z"},"languages":{"value":["python"],"source":"github.language","observed_at":"2026-08-23T06:02:05.228Z"},"license_spdx":{"value":"MIT","source":"github.license","observed_at":"2026-08-23T06:02:05.228Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["For enhancing precision in retrieval-augmented generation tasks needing complex query processing","When using BEIR or similar datasets to benchmark your generation models"],"when_not_to_use":["If you require real-time performance, as multi-query generation may introduce latency","In scenarios where only simple keyword-based search suffices without the need for advanced fusion techniques"],"source":"enrich:decision_facts","observed_at":"2026-07-15T09:54:04.567Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"RAG-Fusion leverages multi-query generation and Reciprocal Rank Fusion for enhanced retrieval-augmented generation tasks, supporting evaluations via NFCorpus/BEIR."}]}}