Qdrant 1.20 ships BM25 hybrid search without sparse vectors
qdrant_engine · x · 2026-10-22
Qdrant 1.20 lets you run hybrid search without the sparse pipeline: enable BM25 scoring directly on a full-text index over payload text and combine it with vector queries in a single request, no embeddings required. The release also adds a code tokenizer for better code search, faster nested filters, and turbo16/turbo8 improvements — a practical update for RAG and search builders.
More from Infra
- Securing coding agents with Docker Sandboxes and edge LLM inference at ATO — HowDevelop · 2026-10-22
- 0.9B Model Transcribes a 4-Person Meeting on a MacBook With 0.08% Speaker Confusion — Binary_orchid · 2026-10-22
- FreeLLM: Pooling Ollama Instances on Employee Laptops for Free Small-Business Inference — JY_Q · 2026-10-22
- Tesla's Terafab aims to build one terawatt of silicon compute per year — yunta_tsai · 2026-10-22
- 40B MoE with 2.5B active params: 4-bit lm_head trick costs only 0.7 pts — AIFlow_ML · 2026-10-22
- With GPU rental prices falling, H100s may need 20 years to pay off — code_star · 2026-10-22