RAG vs Long Context: Why Bigger Context Windows Didn't Kill Retrieval
RAG vs long context, with the arithmetic: per-query cost, prefill latency, effective context length, prompt caching limits, and when each one wins.
33 min readRaj
Verified against NoLiMa (arXiv:2502.05167v3); Self-Route (arXiv:2407.16833); vendor pricing read 2026-09-22