RAG
5 published articles
RAG research
BM25 beats agentic RAG when corpora pass 10 million tokens
Agentic search wins on small corpora, but a muset-ai study across 28 nested tiers shows BM25 overtaking it near 10 million tokens. The agent burns 39x more query tokens, and graph RAG stalls in construction.
2026-08-07
MemTensor's Memory Foundation Models
AI agents kept memory outside the model. Metis puts it inside
MemTensor's Metis is the first prototype of memory foundation models: history compressed into the backbone, read back through memory attention, updated in one forward pass. It challenges the vector-store era of agent memory. The limits are acknowledged but not yet quantified.
2026-07-30
AI Models & Infrastructure
Nvidia just proved that better embeddings pay for themselves in agent runtime
Nvidia's Nemotron 3 Embed collection claims the #1 spot on the RTEB benchmark and introduces 1B variants that retain 99% of the 8B model's accuracy. Company data shows stronger retrieval reduces downstream token costs in agentic systems.
2026-07-18
Open Source Infrastructure
Your AI search pipeline is broken. This open-source framework fixes the plumbing.
Teams building AI search infrastructure still spend too much time on plumbing. Search Toolkit unifies ingestion, retrieval, and evaluation into a single open-source framework, eliminating the weeks of integration work needed to stitch together separate tools. It's designed for enterprise use cases like RAG quality, domain-specific retrieval, and agentic search.
2026-07-11
Tooling
Your AI model is a commodity. The pipeline is where the real advantage lives.
A practical, step-by-step analysis of building an AI writing pipeline in 2025: model selection, prompt chaining, and quality control. No hype, just the technical architecture that matters.
2026-07-11