headpiece747/ninfer-5090-windows

Native Windows port of NInfer engine for RTX 5090. Features Qwen3.8-27B with QUASAR and NInfer models, MTP/DFlash2 with vision and 262,144 context

语言：C++ · GitHub 星标：84 · Fork：14

话题：cuda · inference-engine · llm-inference · msvc · nvfp4 · quantization · qwen3 · rtx-5090

在 GitHub 上查看原仓库