maas-base/mkdocs.yml

# Project information
site_name: MASS-Base
site_url: https://docs.gpustack.ai
site_author: MASS-Base
site_description: >-
  MASS-Base is an open-source GPU cluster manager designed for efficient AI model deployment.
  It lets you run models efficiently on your own GPU hardware by choosing the best inference engines,
  scheduling GPU resources, analyzing model architectures, and automatically configuring deployment parameters.

# Repository
repo_name: gpustack/gpustack
repo_url: https://github.com/gpustack/gpustack

# Copyright
copyright: Copyright © 2026 GPUStack.ai

theme:
  name: material
  logo: assets/logo-white.png
  favicon: assets/logo.png
  custom_dir: docs/overrides
  palette:
    - media: "(prefers-color-scheme)"
      toggle:
        icon: material/brightness-auto
        name: Switch to light mode
    - media: "(prefers-color-scheme: light)"
      scheme: default
      primary: custom
      accent: orange
      toggle:
        icon: material/brightness-7
        name: Switch to dark mode
    - media: "(prefers-color-scheme: dark)"
      scheme: slate
      primary: light blue
      accent: orange
      toggle:
        icon: material/brightness-4
        name: Switch to system preference
  features:
    - search.suggest
    - search.highlight
    - content.tabs.link
    - navigation.indexes
    - content.tooltips
    - navigation.path
    - navigation.tabs
    - content.code.annotate
    - content.code.copy
    - content.code.select
    - content.action.view
    - content.action.edit
    - navigation.top
    - navigation.footer
    - navigation.tracking
    - search.share
    - toc.follow
  icon:
    repo: fontawesome/brands/github
  language: en
plugins:
  - glightbox
  - search
  - redirects:
      redirect_maps:
        'installation/nvidia/installation.md': 'installation/requirements.md'
        'installation/amd/installation.md': 'installation/requirements.md'
        'installation/ascend/installation.md': 'installation/requirements.md'
        'installation/hygon/installation.md': 'installation/requirements.md'
        'installation/mthreads/installation.md': 'installation/requirements.md'
        'installation/iluvatar/installation.md': 'installation/requirements.md'
        'installation/cambricon/installation.md': 'installation/requirements.md'
        'installation/metax/installation.md': 'installation/requirements.md'
  - minify:
      minify_html: true
      minify_css: true
      minify_js: true
      cache_safe: true
      htmlmin_opts:
        remove_comments: true
extra:
  generator: false
  version:
    provider: mike
  social:
    - icon: fontawesome/brands/github
      link: https://github.com/gpustack
    - icon: fontawesome/brands/docker
      link: https://hub.docker.com/r/gpustack
    - icon: fontawesome/brands/python
      link: https://pypi.org/project/gpustack/
    - icon: fontawesome/brands/discord
      link: https://discord.gg/VXYJzuaqwD
    - icon: fontawesome/brands/weixin
      link: https://github.com/gpustack/gpustack/blob/main/docs/assets/wechat-group-qrcode.jpg
extra_javascript:
  - image-selector/assets/extra.js
  - https://unpkg.com/katex@0/dist/katex.min.js
  - javascripts/katex.js
  - https://unpkg.com/katex@0/dist/contrib/auto-render.min.js
  - https://unpkg.com/tablesort@5.3.0/dist/tablesort.min.js
  - javascripts/tablesort.js
extra_css:
  - stylesheets/extra.css
  - https://unpkg.com/katex@0/dist/katex.min.css
markdown_extensions:
  - abbr
  - def_list
  - attr_list
  - md_in_html
  - pymdownx.highlight:
      linenums: true
      line_spans: __span
      anchor_linenums: true
      pygments_lang_class: true
  - pymdownx.inlinehilite
  - pymdownx.superfences:
      custom_fences:
        - name: mermaid
          class: mermaid
          format: !!python/name:pymdownx.superfences.fence_code_format
  - pymdownx.tabbed:
      alternate_style: true
  - admonition
  - sane_lists
  - pymdownx.details
  - pymdownx.tilde
  - pymdownx.betterem
  - pymdownx.caret
  - pymdownx.tasklist:
      custom_checkbox: true
  - pymdownx.arithmatex:
      generic: true
  - pymdownx.emoji:
      emoji_generator: !!python/name:material.extensions.emoji.to_svg
      emoji_index: !!python/name:material.extensions.emoji.twemoji
  - pymdownx.critic
  - pymdownx.keys
  - pymdownx.snippets:
      auto_append:
        - docs/includes/abbreviations.md
edit_uri: edit/main/docs/
nav:
  - Home:
      - Overview: overview.md
      - Quickstart: quickstart.md
      - Installation:
          - Requirements: installation/requirements.md
          - Installation: installation/installation.md
          - Air-Gapped Installation: installation/air-gapped.md
          - Uninstallation: installation/uninstallation.md
      - Upgrade: upgrade.md
      - Migration: migration.md
      - User Guide:
          - Playground:
              - Playgound: user-guide/playground/index.md
              - Chat: user-guide/playground/chat.md
              - Image: user-guide/playground/image.md
              - Audio: user-guide/playground/audio.md
              - Embedding: user-guide/playground/embedding.md
              - Rerank: user-guide/playground/rerank.md
          - Model Catalog: user-guide/model-catalog.md
          - Model Deployment Management: user-guide/model-deployment-management.md
          - Model Route Management: user-guide/model-route-management.md
          - Model Provider Management: user-guide/model-provider-management.md
          - Inference Backend Management: user-guide/inference-backend-management.md
          - Built-in Inference Backends: user-guide/built-in-inference-backends.md
          - Compatibility Check: user-guide/compatibility-check.md
          - Model File management: user-guide/model-file-management.md
          - Cluster Management: user-guide/cluster-management.md
          - Cloud Credential Management: user-guide/cloud-credential-management.md
          - API Key Management: user-guide/api-key-management.md
          - User Management: user-guide/user-management.md
          - Single Sign-On (SSO) Authentication: user-guide/sso.md
          - Observability: user-guide/observability.md
          - Benchmarking: user-guide/benchmarking.md
      - Using Models:
          - Using Large Language Models: using-models/using-large-language-models.md
          - Using Vision Language Models: using-models/using-vision-language-models.md
          - Using Embedding Models: using-models/using-embedding-models.md
          - Using Reranker Models: using-models/using-reranker-models.md
          - Using Image Generation Models: using-models/using-image-generation-models.md
          - Using Audio Models: using-models/using-audio-models.md
      - Tutorials:
          - Running DeepSeek R1 671B with Distributed vLLM: tutorials/running-deepseek-r1-671b-with-distributed-vllm.md
          - Running DeepSeek R1 671B with Distributed Ascend Mindie: tutorials/running-deepseek-r1-671b-with-distributed-ascend-mindie.md
          - Inference On CPUs: tutorials/inference-on-cpus.md
          - Inference with Tool Calling: tutorials/inference-with-tool-calling.md
          - Using Custom Inference Backend: tutorials/using-custom-backends.md
          - Adding a GPU Cluster Using DigitalOcean: tutorials/adding-gpucluster-using-digitalocean.md
          - Adding a GPU Cluster Using Kubernetes: tutorials/adding-gpucluster-using-kubernetes.md
          - Managing Model Routes: tutorials/managing-model-routes.md
      - Integrations:
          - Inference APIs: integrations/inference-apis.md
          - Integrate with Claude Code: integrations/integrate-with-claude-code.md
          - Integrate with Dify: integrations/integrate-with-dify.md
          - Integrate with RAGFlow: integrations/integrate-with-ragflow.md
          - Integrate with CherryStudio: integrations/integrate-with-cherrystudio.md
          - Integrate with OpenClaw: integrations/integrate-with-openclaw.md
          - Integrate with n8n: integrations/integrate-with-n8n.md
          - Integrate with MaxKB: integrations/integrate-with-maxkb.md
      - Architecture: architecture.md
      - Scheduler: scheduler.md
      - Troubleshooting: troubleshooting.md
      - FAQ: faq.md
      - API Reference: api-reference.md
      - CLI Reference:
          - Start: cli-reference/start.md
          - Download Tools: cli-reference/download-tools.md
          - Reload Config: cli-reference/reload-config.md
          - List Images: cli-reference/list-images.md
          - Save Images: cli-reference/save-images.md
          - Load Images: cli-reference/load-images.md
          - Copy Images: cli-reference/copy-images.md
      - Environment Variables: environment-variables.md
  - Inference Performance Lab:
      - Overview: performance-lab/overview.md
      - Optimizing Throughput:
          - Qwen3.5-35B-A3B:
              - H200: performance-lab/qwen3.5-35b-a3b/h200.md
          - Qwen3.5-9B:
              - H100: performance-lab/qwen3.5-9b/h100.md
          - GLM-4.5-Air:
              - A100: performance-lab/glm-4.5-air/a100.md
              - H100: performance-lab/glm-4.5-air/h100.md
          - GLM-4.x:
              - A100: performance-lab/glm-4.x/a100.md
              - H100: performance-lab/glm-4.x/h100.md
              - H200: performance-lab/glm-4.x/h200.md
          - GPT-OSS-20B:
              - A100: performance-lab/gpt-oss-20b/a100.md
              - H100: performance-lab/gpt-oss-20b/h100.md
          - GPT-OSS-120B:
              - A100: performance-lab/gpt-oss-120b/a100.md
              - H100: performance-lab/gpt-oss-120b/h100.md
          - DeepSeek-R1:
              - H200: performance-lab/deepseek-r1/h200.md
          - DeepSeek-V3.2:
              - H200: performance-lab/deepseek-v3.2/h200.md
          - Qwen3-8B:
              - 910B: performance-lab/qwen3-8b/910b.md
          - Qwen3-14B:
              - A100: performance-lab/qwen3-14b/a100.md
              - H100: performance-lab/qwen3-14b/h100.md
          - Qwen3-32B:
              - A100: performance-lab/qwen3-32b/a100.md
              - H100: performance-lab/qwen3-32b/h100.md
          - Qwen3-30B-A3B:
              - 910B: performance-lab/qwen3-30b-a3b/910b.md
          - Qwen3-235B-A22B:
              - A100: performance-lab/qwen3-235b-a22b/a100.md
              - H100: performance-lab/qwen3-235b-a22b/h100.md
      - Optimizing Latency:
          - Qwen3.5-35B-A3B:
              - H200: performance-lab/qwen3.5-35b-a3b/h200-latency.md
          - Qwen3.5-9B:
              - H100: performance-lab/qwen3.5-9b/h100-latency.md
          - Qwen3-8B:
              - H100: performance-lab/qwen3-8b/h100-latency.md
      - References:
          - The Impact of Quantization on vLLM Inference Performance: performance-lab/references/the-impact-of-quantization-on-vllm-inference-performance.md
          - Evaluating LMCache Prefill Acceleration in vLLM: performance-lab/references/evaluating-lmcache-prefill-acceleration-in-vllm.md
  - Container Image Selector: image-selector/index.html