8bit.tr Journal
Attention
2 articles tagged with Attention.
December 31, 2025
Mixture of Attention Routing: Smarter Context Allocation at Scale
A technical exploration of attention routing strategies that allocate context budget to the most relevant tokens.
December 22, 2025
KV Cache and Attention Optimization: The Hidden Performance Layer
A deep technical guide to KV caching, attention optimization, and memory-aware serving for LLMs.