Attemory: A Retrieval Engine Replacing Embeddings with Attention
langsfang · reddit · 2026-07-06
Developer langsfang released Attemory, an open-source retrieval engine that replaces traditional embedding-based nearest neighbor search by indexing raw text into reusable KV states. During retrieval, a local Qwen3.5 model performs attention computation over the indexed memory and the query, returning compact evidence like memory IDs, snippets, or file line ranges. The author argues that attention-based retrieval preserves token-level details lost in embeddings, achieves SOTA on long-memory and code benchmarks, and eliminates the need for stacking tricks like chunk tuning and re-ranking.
More from coding & agent
- Codex adoption in Germany: weekly active users up 20x since January 2026 — testingcatalog · 2026-10-26
- Filtering secrets on the write path: why agent memory deserves more suspicion at ingest than retrieval — Classic-Being-7960 · 2026-10-26
- INT21's AI agent swarm rewrote Drama 3's inference runtime in Rust within one hour, beating SGLang-Omni — bingxu_ · 2026-10-26
- After OpenCode key leak: 5 practical rules for keeping keys away from agents — jacobholm_dev · 2026-10-26
- wandb 0.31.0 ships with 2x faster startup, adds JSON output for coding agents — wandb · 2026-10-26
- $150 modular Linux handheld Mecha Comet opens preorders — pocket SSH and Claude Code included — Paimaamu · 2026-10-26