Published November 12, 2026 | Version v1

Llama vs Qwen vs Gemma on a CPU: 3.5 GiB of KV Cache for a 0.6B Model

Authors/Creators

  • 1. Odessa National Polytechnic University

Description

Research article: Llama vs Qwen vs Gemma on a CPU: 3.5 GiB of KV Cache for a 0.6B Model

Files

3752.md

Files (27.8 kB)

Name Size
md5:906bb3aeb40fd7f0b148970e17486937
27.8 kB Preview Download