The subtle KV cache detail most people get wrong: tokens generate before entering the cache

techNmak · x · 2026-09-18

A clear explanation of a common KV cache misconception in decoder-only Transformers: not every token flows through the LLM the same way.

Key takeaway: a token can be generated before it has been processed into the KV cache — a distinction that matters a lot for understanding decoding.

Original post →

More from Research

Research channel →