deepseek2024-v2

arXiv： 2405.04434

TLDR（中文）

引入 Multi-head Latent Attention (MLA) 把 KV cache 砍到 1/13，让 236B MoE 推理价格碾压同档闭源。MLA 是 V3/R1 推理性价比的核心来源。

TLDR (English)

Introduces Multi-head Latent Attention (MLA) reducing KV cache to 1/13, making 236B MoE inference price crush same-tier closed-source. MLA is core source of V3/R1 inference cost-effectiveness.