002f9879b2
Replaces cherry-picked upstream_ref with complete source trees. xllm/ — Iluvatar official C++ inference engine (15MB, 1470 files) Complete: kernels → layers → models → runtime → scheduler → api Excluded: .git, binary images, third_party submodule checkouts ds_vllm/ — Iluvatar official vllm fork (8MB, 703 files) Included: csrc/ (ALL CUDA kernels), fused_moe/, qwen3_5 model, _custom_ops Excluded: tests, benchmarks, docs, examples (not needed for reference) Critical call chains now fully traceable: MoE: moe_topk_softmax_kernels.cuh → ixformer.h → fused_moe.cpp → layer GDN: qwen3_gated_delta_net_base.cpp → qwen3_5_gated_delta_net.cpp Attention: ixformer.h → xllm_paged_attention → attention.cpp
282 lines
7.5 KiB
YAML
282 lines
7.5 KiB
YAML
site_name: xLLM
|
|
site_url: !ENV READTHEDOCS_CANONICAL_URL
|
|
|
|
# Repository
|
|
repo_name: jd-opensource/xllm
|
|
repo_url: https://github.com/jd-opensource/xllm
|
|
|
|
edit_uri: edit/main/docs/zh
|
|
use_directory_urls: true
|
|
# Copyright
|
|
copyright: Copyright © 2025 xLLM Team
|
|
# docs_dir: docs/zh
|
|
# Configuration
|
|
theme:
|
|
name: material
|
|
language: zh
|
|
custom_dir: docs/mkdocs/overrides
|
|
features:
|
|
- announce.dismiss
|
|
- content.action.edit
|
|
- content.action.view
|
|
- content.code.annotate
|
|
- content.code.copy
|
|
- content.code.select
|
|
- content.footnote.tooltips
|
|
# - content.tabs.link
|
|
- content.tooltips
|
|
# - header.autohide
|
|
- navigation.expand
|
|
- navigation.footer
|
|
- navigation.indexes
|
|
- navigation.path
|
|
# - navigation.instant
|
|
# - navigation.instant.prefetch
|
|
# - navigation.instant.progress
|
|
# - navigation.prune
|
|
- navigation.sections
|
|
- navigation.tabs
|
|
# - navigation.tabs.sticky
|
|
- navigation.top
|
|
- navigation.tracking
|
|
- search.highlight
|
|
- search.share
|
|
- search.suggest
|
|
- toc.follow
|
|
# - toc.integrate
|
|
palette:
|
|
- media: "(prefers-color-scheme)"
|
|
toggle:
|
|
icon: material/link
|
|
name: 切换亮色模式
|
|
- media: "(prefers-color-scheme: light)"
|
|
scheme: default
|
|
primary: custom
|
|
accent: custom
|
|
toggle:
|
|
# icon: material/toggle-switch
|
|
icon: material/eye
|
|
name: 切换暗色模式
|
|
- media: "(prefers-color-scheme: dark)"
|
|
scheme: slate
|
|
primary: black
|
|
accent: custom
|
|
toggle:
|
|
# icon: material/toggle-switch-off
|
|
icon: material/eye-outline
|
|
name: 切换到系统偏好
|
|
# primary: custom
|
|
# accent: custom
|
|
font:
|
|
text: Roboto
|
|
code: Roboto Mono
|
|
favicon: assets/logo.png
|
|
logo: assets/logo.png
|
|
icon:
|
|
annotation: material/arrow-right-circle
|
|
# icon:
|
|
# logo: logo
|
|
# Plugins
|
|
plugins:
|
|
# - awesome-pages
|
|
# - blog
|
|
# - static-i18n
|
|
- git-revision-date-localized:
|
|
exclude:
|
|
- argparse/*
|
|
- examples/*
|
|
- search:
|
|
separator: '[\s\u200b\-_,:!=\[\]()"`/]+|\.(?!\d)|&[lg]t;|(?!\b)(?=[A-Z][a-z])'
|
|
- minify:
|
|
minify_html: true
|
|
extra_css:
|
|
- mkdocs/stylesheets/extra.css
|
|
# extra_dir: assets
|
|
# Hooks
|
|
# hooks:
|
|
# - material/overrides/hooks/shortcodes.py
|
|
# - material/overrides/hooks/translations.py
|
|
|
|
# Additional configuration
|
|
extra:
|
|
source:
|
|
path: https://github.com/jd-opensource/xllm/blob/main/docs/zh/
|
|
|
|
status:
|
|
new: Recently added
|
|
deprecated: Deprecated
|
|
# analytics:
|
|
# provider: google
|
|
# property: !ENV GOOGLE_ANALYTICS_KEY
|
|
social:
|
|
- icon: fontawesome/brands/github
|
|
link: https://github.com/jd-opensource/xllm
|
|
- icon: gitcodeai
|
|
link: https://gitcode.com/xLLM-AI/xllm
|
|
- icon: fontawesome/brands/docker
|
|
link: https://hub.docker.com/r/xllm/xllm-ai
|
|
- icon: email-fill
|
|
link: https://github.com/jd-opensource/xllm
|
|
# - icon: fontawesome/brands/python
|
|
# link: https://pypi.org/project/mkdocs-material/
|
|
# - icon: fontawesome/brands/bluesky
|
|
# link: https://bsky.app/profile/squidfunk.bsky.social
|
|
# - icon: fontawesome/brands/mastodon
|
|
# link: https://fosstodon.org/@squidfunk
|
|
# - icon: fontawesome/brands/x-twitter
|
|
# link: https://x.com/squidfunk
|
|
|
|
# Extensions
|
|
markdown_extensions:
|
|
- tables
|
|
- abbr
|
|
- admonition
|
|
- attr_list
|
|
- def_list
|
|
- footnotes
|
|
- md_in_html
|
|
- toc:
|
|
permalink: true
|
|
- pymdownx.arithmatex:
|
|
generic: true
|
|
- pymdownx.betterem:
|
|
smart_enable: all
|
|
- pymdownx.caret
|
|
- pymdownx.details
|
|
- pymdownx.emoji:
|
|
emoji_generator: !!python/name:material.extensions.emoji.to_svg
|
|
emoji_index: !!python/name:material.extensions.emoji.twemoji
|
|
- pymdownx.highlight:
|
|
anchor_linenums: true
|
|
line_spans: __span
|
|
pygments_lang_class: true
|
|
- pymdownx.inlinehilite
|
|
- pymdownx.keys
|
|
- pymdownx.magiclink:
|
|
normalize_issue_symbols: true
|
|
repo_url_shorthand: true
|
|
user: jd-opensource
|
|
repo: xllm
|
|
- pymdownx.mark
|
|
- pymdownx.smartsymbols
|
|
- pymdownx.snippets
|
|
- pymdownx.superfences
|
|
- pymdownx.tabbed
|
|
# - pymdownx.superfences:
|
|
# custom_fences:
|
|
# - name: mermaid
|
|
# class: mermaid
|
|
# format: !!python/name:pymdownx.superfences.fence_code_format
|
|
# - pymdownx.tabbed:
|
|
# alternate_style: true
|
|
# combine_header_slug: true
|
|
# slugify: !!python/object/apply:pymdownx.slugs.slugify
|
|
# kwds:
|
|
# case: lower
|
|
- pymdownx.tasklist:
|
|
custom_checkbox: true
|
|
- pymdownx.tilde
|
|
- mdx_math:
|
|
enable_dollar_delimiter: true
|
|
extra_javascript:
|
|
- mkdocs/javascripts/mathjax.js
|
|
- https://unpkg.com/mathjax@3/es5/tex-mml-chtml.js
|
|
|
|
|
|
nav:
|
|
- 主页: index.md
|
|
- 用户指南:
|
|
- getting_started/quick_start.md
|
|
- supported_models.md
|
|
- getting_started/launch_xllm.md
|
|
- getting_started/disagg_pd.md
|
|
- getting_started/multi_machine.md
|
|
- 基本用法:
|
|
- getting_started/online_service.md
|
|
- getting_started/offline_service.md
|
|
- 功能介绍:
|
|
- 概览:
|
|
- features/overview.md
|
|
- 功能列表:
|
|
- 运行时:
|
|
- features/async_schedule.md
|
|
- features/multi_streams.md
|
|
- 调度器:
|
|
- features/continuous_scheduler.md
|
|
- features/chunked_scheduler.md
|
|
- features/zero_evict_scheduler.md
|
|
- 分布式:
|
|
- features/disagg_pd.md
|
|
- 算子:
|
|
- features/groupgemm.md
|
|
- features/ppmatmul.md
|
|
- features/topk_topP.md
|
|
- KV存储管理:
|
|
- features/prefix_cache.md
|
|
- features/global_kvcache.md
|
|
- features/xtensor_memory.md
|
|
- 多模态:
|
|
- features/multimodal.md
|
|
- MoE优化:
|
|
- features/moe_params.md
|
|
- features/eplb.md
|
|
- 进阶指南:
|
|
- 投机推理: features/mtp.md
|
|
- GraphMode: features/graph_mode.md
|
|
- xLLM Service概览: features/xllm_service_overview.md
|
|
- 开发者指南:
|
|
- dev_guide/code_arch.md
|
|
- dev_guide/tilelang_ascend_kernel_dev.md
|
|
- CLI参考:
|
|
- cli_reference.md
|
|
|
|
# nav:
|
|
# - 主页: index.md
|
|
# - 开始使用: getting_started/
|
|
# - 功能介绍: features/
|
|
# - CLI参考: cli_reference.md
|
|
# nav:
|
|
# - 主页: zh/index.md
|
|
# - 开始使用:
|
|
# - zh/getting_started/compile.md
|
|
# - 快速开始:
|
|
# - 在线服务:
|
|
# - zh/getting_started/single_node.md
|
|
# - zh/getting_started/multi_node.md
|
|
# - zh/getting_started/PD_disagg.md
|
|
# - 多模态支持:
|
|
# - zh/getting_started/start_vlm_service.md
|
|
# - zh/getting_started/scripts.md
|
|
# - 功能介绍:
|
|
# - 概览:
|
|
# - zh/features/overview.md
|
|
# - 功能列表:
|
|
# - 运行时:
|
|
# - zh/features/async_schedule.md
|
|
# - zh/features/multi_streams.md
|
|
# - 调度算法:
|
|
# - zh/features/continuous_scheduler.md
|
|
# - zh/features/chunked_scheduler.md
|
|
# - zh/features/zero_evict_scheduler.md
|
|
# - zh/features/prefix_cache.md
|
|
# - 分布式:
|
|
# - zh/features/disagg_pd.md
|
|
# - 算子:
|
|
# - zh/features/groupgemm.md
|
|
# - zh/features/ppmatmul.md
|
|
# - zh/features/topk_topP.md
|
|
# - KV存储管理:
|
|
# - zh/features/global_kvcache.md
|
|
# - zh/features/xtensor_memory.md
|
|
# - 多模态:
|
|
# - zh/features/multimodal.md
|
|
# - MoE优化:
|
|
# - zh/features/moe_params.md
|
|
# - zh/features/eplb.md
|
|
# - 进阶指南:
|
|
# - 投机推理: zh/features/mtp.md
|
|
# - ACLGraph: zh/features/acl_graph.md
|
|
# - xLLM Service概览: zh/features/xllm_service_overview.md
|
|
# - CLI参考:
|
|
# - zh/cli_reference.md
|