hybrid-rag · service and data flow

BM25 + dense relative-score fusion, optional TEI reranking, then bounded generation

hybrid-rag service and data flowAtlas ingestion into RagBase followed by fused keyword and vector retrieval, optional TEI reranking, and answer generation. ATLAS COMPOSE STACK · RAG TRACKSHARED CORPUS PREPARATIONQUERY PATH Corpus documentsselected dataset/profile Atlas ingestionparse · chunk · metadata Embedding modelnomic-embed-text Weaviate · RagBasetext index + vectors CallerOpen WebUI / API LiteLLMmodel: hybrid-rag Hybrid pluginfusion · prompt · metrics Query embeddingnomic-embed-text Weaviate hybridBM25 + dense · α=0.5 TEI rerankercross-encoder · optional light_genquestion + top context OpenAI responseanswer · sources · metrics 1request 2dispatch 3embed 4vector 520 candidates 6top 5 7answer rerank=false · bypass TEI and preserve fused order MESSAGE CONTRACT AND TUNING SURFACES 0 · INDEXRagBase stores lexical text and dense vectors.The same index supports both retrieval signals.No graph is constructed. 1–4 · RETRIEVAL REQUESTquestion + vector → Weaviate hybrid queryalpha balances BM25 and dense relevance.retrieve_k controls the candidate pool. 5–6 · OPTIONAL RERANKquery + candidate texts → TEI /reranktop_n bounds the generation context.Disabled mode bypasses TEI explicitly. 7 · GENERATE + RETURNlight_gen receives ordered evidence.Response preserves source identifiers and scores.Metrics expose retrieval and generation latency.