lxe/dsx

DeepSeek-V4.1-Flash-1017 (552B MoE + 196B Engram) on 2x RTX 3090 + 128 GB RAM. Rust + CUDA, T16 quantization. 40+ tok/s measured with expert deferral + FP8 projections.

语言：Rust · GitHub 星标：88 · Fork：19

在 GitHub 上查看原仓库