hygon 开源项目推荐（14 个）

Project-HAMi /HAMi

Kubernetes 上的异构 GPU 共享

FlyAIBox /dcu-in-action

国产加速卡-海光DCU实战（大模型训练、微调、推理 等）

nightthink /K100AI-Model-Deploy

K100AI 模型部署实践 —— 海光 K100-AI 上大模型的成功配置线拉起包（当前：Qwen3.8-27B 六条精选线；持续更新）

DocPang /qwen36-k100ai-w8a8-optimization

在 vLLM 0.18.1 上针对 Hygon K100AI（gfx928）的可复现单 GPU Qwen3.6-35B-A3B W8A8 推理调优。

Entropic-Silence /unsloth-studio-k100-dtk2604

Unsloth Studio deployment for Hygon K100-aiDCU on DTK 26.04

Hong-agent /K100LC-RT4

海光 K100 LC（gfx926）上从零自研的 Qwen3.8-27B 推理运行时：RT4 权重格式、int4 算子、MTP3 投机解码、本地视觉塔与 OpenAI 兼容服务

lql341 /dsh-scnet

dsh plugin for scnet.cn

837515265 /llm-serving-platform

Open-source model serving and heterogeneous GPU management platform for enterprise private AI. NVIDIA / Ascend / Cambricon / Hygon, unified through OpenAI-compatible APIs.

DocPang /minimax-h3-k100ai-optimization

MiniMax H3 INT8 ConvRot optimization and dual-K100AI QKV/Attention inference for Hygon K100AI (gfx928)

970334745 /deepseek-v4-hygon-bw100

Scoped DeepSeek V4 paged-MQA correctness repair for Hygon BW100 (gfx936), TP8 integration, and historical million-token validation.

lql341 /scnet-app

Meta-skill for integrating HPC, AI4S and scientific applications with SCNet compute resources, Slurm scheduling and cluster packaging, including Hygon DCU compatibility.

mcocdaa /hysmi-top

Terminal monitor for Hygon DCU (hy-smi) with nvtop-style scrolling curves

mengshi02 /nano-vllm-hal

A miniature of vLLM v1 with a vLLM-style Platform/op-family hardware abstraction layer. One engine, four backends: NVIDIA CUDA, Hygon DCU (DTK/HIP), Ascend NPU, Moore Threads MUSA. Real continuous batching, paged KV cache, TP, CUDA graph — runs Qwen3/Olmoe end-to-end

lql341 /scnet-hpc

Agent skill for working on SCNet clusters