2026-11-02 · America/Los_Angeles · 社区动态 · #18

llama + CUDA: flash_attn_ext_rows (fix unified kv for multi-seq) by am17an · Pull Request #29510 · ggml-org/llama.cpp

Do you run multiple agents locally like I do? If so, this is a really nice speedup for the unified KV cache

热度 44.4 / 100；排名与评分保留该期记录。