Originally published on rohitraj.tech
A 1M-token context window is not memory — it is RAM that gets wiped when the session ends. On the LoCoMo benchmark a two-layer memory setup hit 91.6% accuracy at ~6,956 tokens vs 72.9% at ~26,000 tokens for full context. Here is the builder read: why context windows behave like RAM, the two-layer architecture every production agent needs, a minimal pgvector memory layer in code, and when a bigger window is still the right call.
Read the full version with code samples, diagrams, and architecture details: AI Agent Memory vs Context Window: Why a Bigger Window Isn't Memory (2026)
More engineering notes: rohitraj.tech/en/notes
Top comments (0)