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 14, 05:30 UTC
[Docs] Improve API docs (+small tweaks) (#22459)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Harry Mellor committed
Aug 8, 2025 at 11:02 UTC
7be7f3824a2d610299991ceefb1b034b3a923b0f
5 files changed
+5
-13
docs/.nav.yml
+2
-3
@@ -58,10 +58,9 @@ nav:
58
- CI: contributing/ci
59
- Design Documents: design
60
- API Reference:
61
- - Summary: api/README.md
61
+ - Summary: api/summary.md
62
- Contents:
63
- - glob: api/vllm/*
64
- preserve_directory_names: true
63
+ - api/vllm/*
64
- CLI Reference:
65
- Summary: cli/README.md
66
- Community:
docs/api/summary.md
renamed
docs/features/quantization/inc.md
+1
-4
@@ -1,7 +1,4 @@
1
----
2
-title: FP8 INC
3
----
4
-[](){ #inc }
1
+# FP8 INC
2
3
vLLM supports FP8 (8-bit floating point) weight and activation quantization using Intel® Neural Compressor (INC) on Intel® Gaudi® 2 and Intel® Gaudi® 3 AI accelerators.
4
Currently, quantization is validated only in Llama models.
docs/mkdocs/hooks/generate_examples.py
+1
-1
@@ -105,7 +105,7 @@ class Example:
105
return fix_case(self.path.stem.replace("_", " ").title())
106
107
def generate(self) -> str:
108
- content = f"---\ntitle: {self.title}\n---\n\n"
108
+ content = f"# {self.title}\n\n"
109
content += f"Source <gh-file:{self.path.relative_to(ROOT_DIR)}>.\n\n"
110
111
# Use long code fence to avoid issues with
mkdocs.yaml
+1
-5
@@ -40,6 +40,7 @@ theme:
40
- navigation.sections
41
- navigation.prune
42
- navigation.top
43
+ - navigation.indexes
44
- search.highlight
45
- search.share
46
- toc.follow
@@ -51,11 +52,6 @@ hooks:
52
- docs/mkdocs/hooks/generate_argparse.py
53
- docs/mkdocs/hooks/url_schemes.py
54
54
-# Required to stop api-autonav from raising an error
55
-# https://github.com/tlambert03/mkdocs-api-autonav/issues/16
56
-nav:
57
- - api
58
-
55
plugins:
56
- meta
57
- search