skills.rest
dongg622
dongg622COMMUNITY
Explore all skills by this author →
GitHub

hygon-vllm

Deploy large language models on Hygon DCU hardware with Docker.

Updated Aug 27, 2026
Search Tags:#performance#docker#model scaling#distributed inference#large models#hygon#GPU deployment
One-click install
npx skills add https://github.com/dongg622/china-ai-chip-skill --skill hygon-vllm
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hygon-vllm
Source: https://github.com/dongg622/china-ai-chip-skill/tree/main/Hygon/hygon-vllm
Command: npx skills add https://github.com/dongg622/china-ai-chip-skill --skill hygon-vllm

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and scripts (resource) components.

What problem does it solve?

This Skill streamlines deploying and managing large language models on Hygon DCU GPUs for high-performance inference tasks.

Core Features & Use Cases

  • Distributed Model Deployment: Guides users through setting up distributed inference services supporting models like Qwen, DeepSeek, and LLaMA on Hygon hardware.
  • Model Optimization and Scaling: Facilitates multi-GPU and multi-node deployment, optimizing for performance with tensor and pipeline parallelism.
  • Use Case: An AI engineer wants to deploy a 70B parameter LLaMA model across multiple Hygon GPUs for real-time inference in a production environment.

Quick Start

Use the Hygon vLLM guide to set up distributed model deployment with Docker, configuring environment variables, and starting inference servers efficiently.

Frequently Asked Questions about hygon-vllm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deploy distributed large model inference on Hygon GPUs?▼

Distributed large model inference on Hygon GPUs is deployed using Docker to configure environment variables and start optimized inference servers. The Skill provides detailed instructions for multi-GPU and multi-node setups.

What's the best way to scale large language models like Qwen and DeepSeek on Hygon DCU hardware?▼

Scaling large language models on Hygon DCU hardware is handled through tensor and pipeline parallelism. This approach optimizes performance across multi-GPU configurations for real-time production inference.

Do I need Docker to run multi-GPU inference with vLLM on Hygon DCU?▼

Docker is required to run multi-GPU inference with vLLM on Hygon DCU. You must use it to tune environment variables and execute model-specific setups for efficient operation in production environments.

Can I deploy a 70B parameter LLaMA model across multiple Hygon GPUs for real-time inference?▼

Deploying a 70B parameter LLaMA model across multiple Hygon GPUs for real-time inference is fully supported. The Skill facilitates this by guiding multi-GPU deployment with optimized configurations.

Why does distributed inference on Hygon hardware require environment variable tuning?▼

Distributed inference on Hygon hardware requires environment variable tuning to optimize model-specific setups and hardware utilization. This ensures efficient operation and high-speed performance during production workloads.

Related Skills

Explore More Software Engineering Skills→
disusereddisusered

jira

Manage Jira work items and comments via the acli command-line tool.

Community
Intermediate
dalagerdalager
1

V3 MCP Optimization

Optimize claude-flow v3 MCP server architecture for reduced latency and throughput.

Community
Advanced
ovachieverovachiever
52

serving-llms-vllm

Deploy OpenAI-compatible LLM APIs with vLLM for high-throughput inference.

Community
Advanced
jonnabiojonnabio
6

openrlhf-training

Train large language models with RLHF algorithms using distributed GPU resources.

Community
Advanced