Radar / AI infrastructure / BRIDGE

BRIDGE: minimal RAG memory for agents

A deliberately minimal RAG memory for agents (chained observation notes, BM25+embedding retrieval, chronological rendering) that matches state-of-the-art agent memory systems on LongMemEval-S and BEAM and beats the best memory baseline on EverMemBench by 5.7–6.3 points.

new launchinfra · memoryJingyu6/rag-in-a-trenchcoat · 3 ★added 2026-11-19
Open BRIDGE →View on the radar

Why it matters

Shows heavy memory machinery (extraction, reflection, graph synthesis) is often unnecessary; minimal RAG matches SOTA at ~5–7K tokens per query and one LLM call per ingestion or query.

What you could build with it

A solo developer building a long-running research agent could give it BRIDGE as its memory layer — cheap chronological notes instead of a complex graph database — and keep multi-week projects coherent without ballooning token costs.

Does it hold up?

Too early to judge: a two-day-old paper with a 2-star repo; the benchmark claims are the authors' own, but the code is public and the design is simple enough to test in an afternoon.

Built with BRIDGE

Learn more

paper (arXiv) →

First spotted on article: source.

More AI infrastructure

vLLM v0.34.0: extensible KV cache by default, Model Runner V2 only, and 10 new modelsThe open inference engine's latest release spans 1,061 commits from 412 contributors, adds 10 models including Mistral…infra · JEV 0.86llama.cpp v1.0.0: the local-AI engine's first stable releasellama.cpp and ggml hit v1.0.0, promising semantic versioning so breaking C API changes only arrive with major bumps.infra · JEV 0.84Graphify: turn any codebase into a knowledge graph your agent queriesA CLI plus agent skill that parses code with tree-sitter into a queryable knowledge graph of functions, classes…infra · JEV 0.77FreeLLMAPI: one free endpoint across 87 LLM providersA self-hosted router that puts 87 free LLM providers and 1,313 free model endpoints behind one OpenAI-compatible /v1…infra · JEV 0.73DeepSeek App API: the whole consumer app, now as developer APIsDeepSeek opened its app's capabilities through its API: web search, file parsing, URL reading, safety checks, voice…infra · JEV 0.67InferSSD: run 800B-class MoE models from an SSD on consumer hardwareWiCi's open-source inference engine streams MoE expert weights from NVMe so models larger than RAM or VRAM can run…infra · JEV 0.64

Get the week's best AI launches, plus 3 ideas worth building

One email every Saturday. Ranked by traction, not hype. Free.