Popular repositories Loading

1. Strata-2080ti22x2 Strata-2080ti22x2 Public

   Strata (Qwen3.8-Flash-Next) on 2x RTX 2080 Ti 22G (sm_75): decode optimizations (#646), dual-user batch slots (#559), 262144 context

   C++ 2

2. fastllm fastllm Public

   Forked from ztxz16/fastllm

   fastllm是后端无依赖的高性能大模型推理库。同时支持张量并行推理稠密模型和混合模式推理MOE模型，任意10G以上显卡即可推理满血DeepSeek。双路9004/9005服务器+单显卡部署DeepSeek满血满精度原版模型，单并发20tps；INT4量化模型单并发30tps，多并发可达60+。

   C++

3. project-maya project-maya Public

   Forked from mw00/project-maya

   Project Maya - run GLM-5.3-Flash (321B MoE), DeepSeek-V4.1-Flash-1017 (748B MoE) and MiniMax-M3.1-Flash (428B MoE) privately on your own GPU(s), from one card to 16, NVIDIA or AMD, Linux or Windows…

   Cuda