deepseek-v4.1IMAGE

llama.cpp adds native DeepSeek V4.1 Flash support for local inference

Native llama.cpp support removes the third-party patch; rough memory math puts 1-bit quants on 128GB machines, 3-bit on a 256GB Mac Studio, and 4-bit-plus at just over 300GB.

鸡骑猫

@mspd4100Archived 6 days agoOriginally posted 2026.10.25

llama.cpp adds native DeepSeek V4.1 Flash support for local inference

鸡骑猫 @mspd4100Originally posted 2026.10.25

Open original on X

等等，DeepSeek V4.1 Flash 能直接用 llama.cpp 本地跑了！！！原生支持，不用再打第三方补丁那如果搞台 RTX Spark 或者 Mac Studio 呢？ · 128GB 的 RTX Spark、DGX Spark、MacBook：能跑 1-bit 量化 · 256GB 的 Mac Studio：能上 3-bit · 4-bit 以上要 300GB 出头

x.com/mspd4100/status/2114183477205893179Archived 2026.10.25 12:00