@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / decc244767

Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 10, 21:30 UTC

[Docs] Use relative `md` links instead of absolute `html` links for cross referencing (#31494)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

Harry Mellor committed Dec 29, 2025 at 13:33 UTC decc244767f276fdb4ab147ef227835c3a91cd13

5 files changed +12 -14

docs/deployment/docker.md

+2 -2

@@ -84,10 +84,10 @@ DOCKER_BUILDKIT=1 docker build . \

84 If you have not changed any C++ or CUDA kernel code, you can use precompiled wheels to significantly reduce Docker build time.

86 * **Enable the feature** by adding the build argument: `--build-arg VLLM_USE_PRECOMPILED="1"`.

87 - * **How it works**: By default, vLLM automatically finds the correct wheels from our [Nightly Builds](https://docs.vllm.ai/en/latest/contributing/ci/nightly_builds/) by using the merge-base commit with the upstream `main` branch.

87 + * **How it works**: By default, vLLM automatically finds the correct wheels from our [Nightly Builds](../contributing/ci/nightly_builds.md) by using the merge-base commit with the upstream `main` branch.

88 * **Override commit**: To use wheels from a specific commit, provide the `--build-arg VLLM_PRECOMPILED_WHEEL_COMMIT=<commit_hash>` argument.

90 - For a detailed explanation, refer to the documentation on 'Set up using Python-only build (without compilation)' part in [Build wheel from source](https://docs.vllm.ai/en/latest/contributing/ci/nightly_builds.html#precompiled-wheels-usage), these args are similar.

90 + For a detailed explanation, refer to the documentation on 'Set up using Python-only build (without compilation)' part in [Build wheel from source](../contributing/ci/nightly_builds.md#precompiled-wheels-usage), these args are similar.

92 ## Building for Arm64/aarch64

docs/design/debug_vllm_compile.md

+1 -1

@@ -33,7 +33,7 @@ goals while minimizing impact to performance and also helps us (vLLM) when you o

33 For more details on the design, please see the following resources:

35 - [Introduction to vLLM-torch.compile blogpost](https://blog.vllm.ai/2025/08/20/torch-compile.html)

36 -- [vLLM-torch.compile integration design](https://docs.vllm.ai/en/latest/design/torch_compile.html)

36 +- [vLLM-torch.compile integration design](./torch_compile.md)

37 - [vLLM Office Hours #26](https://www.youtube.com/live/xLyxc7hxCJc?si=Xulo9pe53C6ywf0V&t=561)

38 - [Talk at PyTorch Conference 2025](https://youtu.be/1wV1ESbGrVQ?si=s1GqymUfwiwOrDTg&t=725)

docs/features/custom_logitsprocs.md

+1 -1

@@ -180,7 +180,7 @@ The `DummyLogitsProcessor.update_state()` implementation maintains a "sparse" re

180

181 ### Wrapping an Existing Request-Level Logits Processor

182

183 -Although the vLLM engine applies logits processors at batch granularity, some users may want to use vLLM with a "request-level" logits processor implementation - an implementation which operates on individual requests. This will be especially true if your logits processor was developed for vLLM version 0, which required it to be a `Callable` (as described [here](https://docs.vllm.ai/en/v0.10.1.1/api/vllm/logits_process.html)) conforming to the following type annotation:

183 +Although the vLLM engine applies logits processors at batch granularity, some users may want to use vLLM with a "request-level" logits processor implementation - an implementation which operates on individual requests. This will be especially true if your logits processor was developed for vLLM version 0, which required it to be a `Callable` (as described [here][vllm.logits_process]) conforming to the following type annotation:

184

185 ``` python

186 RequestLogitsProcessor = Union[

docs/getting_started/installation/cpu.md

+4 -4

@@ -172,13 +172,13 @@ Note, it is recommended to manually reserve 1 CPU for vLLM front-end process whe

172

173 ### What are supported models on CPU?

174

175 -For the full and up-to-date list of models validated on CPU platforms, please see the official documentation: [Supported Models on CPU](https://docs.vllm.ai/en/latest/models/hardware_supported_models/cpu)

175 +For the full and up-to-date list of models validated on CPU platforms, please see the official documentation: [Supported Models on CPU](../../models/hardware_supported_models/cpu.md)

176

177 ### How to find benchmark configuration examples for supported CPU models?

178

179 -For any model listed under [Supported Models on CPU](https://docs.vllm.ai/en/latest/models/hardware_supported_models/cpu), optimized runtime configurations are provided in the vLLM Benchmark Suite’s CPU test cases, defined in [cpu test cases](https://github.com/vllm-project/vllm/blob/main/.buildkite/performance-benchmarks/tests/serving-tests-cpu.json)

180 -For details on how these optimized configurations are determined, see: [performance-benchmark-details](https://github.com/vllm-project/vllm/tree/main/.buildkite/performance-benchmarks#performance-benchmark-details).

181 -To benchmark the supported models using these optimized settings, follow the steps in [running vLLM Benchmark Suite manually](https://docs.vllm.ai/en/latest/contributing/benchmarks/#manually-trigger-the-benchmark) and run the Benchmark Suite on a CPU environment.

179 +For any model listed under [Supported Models on CPU](../../models/hardware_supported_models/cpu.md), optimized runtime configurations are provided in the vLLM Benchmark Suite’s CPU test cases, defined in [cpu test cases](../../../.buildkite/performance-benchmarks/tests/serving-tests-cpu.json)

180 +For details on how these optimized configurations are determined, see: [performance-benchmark-details](../../../.buildkite/performance-benchmarks/README.md#performance-benchmark-details).

181 +To benchmark the supported models using these optimized settings, follow the steps in [running vLLM Benchmark Suite manually](../../benchmarking/dashboard.md#manually-trigger-the-benchmark) and run the Benchmark Suite on a CPU environment.

182

183 Below is an example command to benchmark all CPU-supported models using optimized configurations.

184

docs/mkdocs/hooks/url_schemes.py

+4 -6

@@ -34,9 +34,10 @@ TITLE = r"(?P<title>[^\[\]<>]+?)"

34 REPO = r"(?P<repo>.+?/.+?)"

35 TYPE = r"(?P<type>issues|pull|projects)"

36 NUMBER = r"(?P<number>\d+)"

37 +PATH = r"(?P<path>[^\s]+?)"

38 FRAGMENT = r"(?P<fragment>#[^\s]+)?"

39 URL = f"https://github.com/{REPO}/{TYPE}/{NUMBER}{FRAGMENT}"

39 -RELATIVE = r"(?!(https?|ftp)://|#)(?P<path>[^\s]+?)"

40 +RELATIVE = rf"(?!(https?|ftp)://|#){PATH}{FRAGMENT}"

42 # Common titles to use for GitHub links when none is provided in the link.

43 TITLES = {"issues": "Issue ", "pull": "Pull Request ", "projects": "Project "}

@@ -55,6 +56,7 @@ def on_page_markdown(

56 title = match.group("title")

57 path = match.group("path")

58 path = (Path(page.file.abs_src_path).parent / path).resolve()

59 + fragment = match.group("fragment") or ""

61 # Check if the path exists and is outside the docs dir

62 if not path.exists() or path.is_relative_to(DOC_DIR):

@@ -64,7 +66,7 @@ def on_page_markdown(

66 slug = "tree/main" if path.is_dir() else "blob/main"

68 path = path.relative_to(ROOT_DIR)

67 - url = f"https://github.com/vllm-project/vllm/{slug}/{path}"

69 + url = f"https://github.com/vllm-project/vllm/{slug}/{path}{fragment}"

70 return f"[{gh_icon} {title}]({url})"

72 def replace_github_link(match: re.Match) -> str:

@@ -88,8 +90,4 @@ def on_page_markdown(

91 markdown = relative_link.sub(replace_relative_link, markdown)

92 markdown = github_link.sub(replace_github_link, markdown)

91 -

92 - if "interface" in str(page.file.abs_src_path):

93 - print(markdown)

94 -

93 return markdown