deepseek-v4.1IMAGE
llama.cpp mainline merges native DeepSeek-V4.1-Flash support
The llama.cpp mainline now runs DeepSeek-V4.1-Flash natively instead of via third-party patches, prompting debate over waiting for quant packages or building from main.
Isaac
@JkwuRmslArchived 6 days agoOriginally posted 2026.10.25
Isaac @JkwuRmslOriginally posted 2026.10.25
llama.cpp 主干刚原生吃上 DeepSeek-V4.1-Flash 以前靠第三方补丁才能跑 本地推理又少折腾一层 你们还在等量化包还是直接上主干了 https://github.com/ggml-org/llama.cpp
x.com/jkwurmsl/status/2114181515068485726Archived 2026.10.25 12:00