Back to Details

model-architecture-diagram

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Model Architecture Diagram

模型架构图

Workflow

工作流程

Return only public original diagrams indexed by this skill.
  1. Run the bundled resolver:
bash
python3 skills/model-architecture-diagram/scripts/model_architecture_diagram.py "<model name>"
  1. If the resolver returns 
    kind: existing
    , return the raw image Markdown it prints and preserve the source attribution line.
  2. If the resolver returns 
    kind: no_match
    , tell the user that no public original architecture diagram is indexed for that model.
仅返回本技能收录的公开原始图表。
  1. 运行内置的解析器：
bash
python3 skills/model-architecture-diagram/scripts/model_architecture_diagram.py "<model name>"
  1. 如果解析器返回
    kind: existing
    ，返回其输出的原始图片Markdown，并保留来源归属行。
  2. 如果解析器返回
    kind: no_match
    ，告知用户该模型暂无收录的公开原始架构图。

Source Priority

源优先级

Use 
references/diagram-index.json
as the source of truth. It stores raw GitHub image URLs from:
  • datawhalechina/self-llm
  • CalvinXKY/InfraTech
  • Tongyi-MAI/Z-Image
  • Wan-Video/Wan2.1
  • Wan-Video/Wan2.2
  • Tencent-Hunyuan/HunyuanVideo
  • Tencent-Hunyuan/Hunyuan3D-2
  • brayevalerien/Flux.1-Architecture-Diagram
Prefer detailed implementation, cookbook, or architecture-card diagrams over paper figures. Good sources show module boundaries, dataflow, MoE / attention / cache paths, or model-specific runtime structure rather than only a high-level paper overview. Official repository diagrams and curated implementation diagrams are first choice; paper figures are fallback only when no more detailed public original diagram is indexed.
Do not copy remote image binaries into the skill. Return the raw GitHub URLs so the chat renderer can display the original image.
以
references/diagram-index.json
为权威数据源，其中存储了来自以下仓库的GitHub原始图片URL：
  • datawhalechina/self-llm
  • CalvinXKY/InfraTech
  • Tongyi-MAI/Z-Image
  • Wan-Video/Wan2.1
  • Wan-Video/Wan2.2
  • Tencent-Hunyuan/HunyuanVideo
  • Tencent-Hunyuan/Hunyuan3D-2
  • brayevalerien/Flux.1-Architecture-Diagram
优先选择详细的实现图、指南图或架构卡片图，而非论文插图。优质数据源应展示模块边界、数据流、MoE/注意力/缓存路径或模型特定的运行时结构，而非仅提供高层级的论文概览。官方仓库图表和精选实现图表为首选；仅当无更详细的公开原始图表收录时，才使用论文插图作为备选。
请勿将远程图片二进制文件复制到技能中。返回原始GitHub URL，以便聊天渲染器显示原始图片。

Existing Diagram Rule

现有图表规则

For a direct match, show the original image. Good direct matches include:
  • DeepSeek V3/V3.2/V4, GLM-5, Kimi K2/K2.5, MiniMax M2.5, Qwen3.5, Qwen3-VL, and Step 3.5 Flash from InfraTech.
  • Hunyuan-A13B, Kimi-VL, Qwen3, Qwen3-VL detail flows, MiniMax M2, and Llama 4 architecture/module diagrams from self-llm.
  • Z-Image, Wan2.1, Wan2.2, HunyuanVideo, Hunyuan3D 2.0, and FLUX.1 diffusion architecture/module diagrams from public GitHub sources.
If multiple diagrams match, show all high-confidence matches up to the resolver's default limit. For example, DeepSeek V3 may return the full architecture plus MLA MHA/MQA diagrams.
对于直接匹配的模型，展示原始图片。优质直接匹配案例包括：
  • InfraTech提供的DeepSeek V3/V3.2/V4、GLM-5、Kimi K2/K2.5、MiniMax M2.5、Qwen3.5、Qwen3-VL和Step 3.5 Flash图表。
  • self-llm提供的Hunyuan-A13B、Kimi-VL、Qwen3、Qwen3-VL详细流程图、MiniMax M2和Llama 4架构/模块图。
  • 公开GitHub源提供的Z-Image、Wan2.1、Wan2.2、HunyuanVideo、Hunyuan3D 2.0和FLUX.1扩散架构/模块图。
若存在多个匹配图表，展示所有高置信度匹配结果，直至解析器的默认限制。例如，DeepSeek V3可能返回完整架构图加MLA MHA/MQA图表。

Hosted Original Diagram Gallery

托管原始图表图库

Do not commit the 
sgl-cookbook-model-architecture-images/
gallery into the repository. The public-original image set is hosted as a GitHub Release asset and indexed by a GitHub issue.
Current hosted artifact:
The artifact contains 44 public original diagram image files from the indexed upstream repositories, plus a lightweight 
index.html
, 
index.md
, 
manifest.json
, HTML contact sheet, and 
architecture-audit.md
.
To inspect the gallery locally:
bash
curl -L -o /tmp/sgl-cookbook-model-architecture-images-2026-05-02.zip \
  https://github.com/BBuf/AI-Infra-Auto-Driven-SKILLS/releases/download/sgl-cookbook-architecture-images-2026-05-02/sgl-cookbook-model-architecture-images-2026-05-02.zip
unzip -q /tmp/sgl-cookbook-model-architecture-images-2026-05-02.zip -d /tmp
open /tmp/sgl-cookbook-model-architecture-images/index.html
请勿将
sgl-cookbook-model-architecture-images/
图库提交到仓库中。公开原始图片集作为GitHub Release资产托管，并通过GitHub Issue建立索引。
当前托管工件：
该工件包含来自索引上游仓库的44个公开原始图表图片文件，以及轻量型的
index.html
、
index.md
、
manifest.json
、HTML联系表和
architecture-audit.md
。
如需本地查看图库：
bash
curl -L -o /tmp/sgl-cookbook-model-architecture-images-2026-05-02.zip \
  https://github.com/BBuf/AI-Infra-Auto-Driven-SKILLS/releases/download/sgl-cookbook-architecture-images-2026-05-02/sgl-cookbook-model-architecture-images-2026-05-02.zip
unzip -q /tmp/sgl-cookbook-model-architecture-images-2026-05-02.zip -d /tmp
open /tmp/sgl-cookbook-model-architecture-images/index.html

Useful Commands

实用命令

List known original diagram aliases:
bash
python3 skills/model-architecture-diagram/scripts/model_architecture_diagram.py --list-known
Emit JSON for automation:
bash
python3 skills/model-architecture-diagram/scripts/model_architecture_diagram.py "GLM-5" --format json
列出已知的原始图表别名：
bash
python3 skills/model-architecture-diagram/scripts/model_architecture_diagram.py --list-known
输出JSON格式用于自动化：
bash
python3 skills/model-architecture-diagram/scripts/model_architecture_diagram.py "GLM-5" --format json

References

参考资料

  • references/diagram-index.json
    : original diagram link index and aliases.
  • references/source-notes.md
    : audited source repositories and local cache paths.
  • references/diagram-index.json
    ：原始图表链接索引及别名。
  • references/source-notes.md
    ：已审核的源仓库及本地缓存路径。