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 13, 04:30 UTC
[Docs] Enable some more markdown lint rules for the docs (#28731)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Harry Mellor committed
Nov 14, 2025 at 18:39 UTC
67187554dd478ba76e79d7a6f8bf02be01290de3
6 files changed
+10
-15
.markdownlint.yaml
-3
@@ -6,9 +6,6 @@ MD024:
6
MD031:
7
list_items: false
8
MD033: false
9
-MD045: false
9
MD046: false
11
-MD051: false
10
MD052: false
13
-MD053: false
11
MD059: false
docs/contributing/benchmarks.md
-2
@@ -10,8 +10,6 @@ vLLM provides comprehensive benchmarking tools for performance testing and evalu
10
- **[Parameter sweeps](#parameter-sweeps)**: Automate `vllm bench` runs for multiple configurations
11
- **[Performance benchmarks](#performance-benchmarks)**: Automated CI benchmarks for development
12
13
-[Benchmark CLI]: #benchmark-cli
14
-
13
## Benchmark CLI
14
15
This section guides you through running benchmark tests with the extensive
docs/contributing/ci/update_pytorch_version.md
+1
-1
@@ -95,7 +95,7 @@ when manually triggering a build on Buildkite. This branch accomplishes two thin
95
to warm it up so that future builds are faster.
96
97
<p align="center" width="100%">
98
- <img width="60%" src="https://github.com/user-attachments/assets/a8ff0fcd-76e0-4e91-b72f-014e3fdb6b94">
98
+ <img width="60%" alt="Buildkite new build popup" src="https://github.com/user-attachments/assets/a8ff0fcd-76e0-4e91-b72f-014e3fdb6b94">
99
</p>
100
101
## Update dependencies
docs/deployment/frameworks/chatbox.md
+2
-2
@@ -29,8 +29,8 @@ pip install vllm
29
- API Path: `/chat/completions`
30
- Model: `qwen/Qwen1.5-0.5B-Chat`
31
32
- 
32
+ 
33
34
1. Go to `Just chat`, and start to chat:
35
36
- 
36
+ 
docs/deployment/frameworks/dify.md
+3
-3
@@ -46,12 +46,12 @@ And install [Docker](https://docs.docker.com/engine/install/) and [Docker Compos
46
- **Model Name for API Endpoint**: `Qwen/Qwen1.5-7B-Chat`
47
- **Completion Mode**: `Completion`
48
49
- 
49
+ 
50
51
1. To create a test chatbot, go to `Studio → Chatbot → Create from Blank`, then select Chatbot as the type:
52
53
- 
53
+ 
54
55
1. Click the chatbot you just created to open the chat interface and start interacting with the model:
56
57
- 
57
+ 
docs/design/fused_moe_modular_kernel.md
+4
-4
@@ -19,9 +19,9 @@ The input activation format completely depends on the All2All Dispatch being use
19
20
The FusedMoE operation is generally made of multiple operations, in both the Contiguous and Batched variants, as described in the diagrams below
21
22
-
22
+
23
24
-
24
+
25
26
!!! note
27
The main difference, in terms of operations, between the Batched and Non-Batched cases is the Permute / Unpermute operations. All other operations remain.
@@ -57,7 +57,7 @@ The `FusedMoEModularKernel` acts as a bridge between the `FusedMoEPermuteExperts
57
The `FusedMoEPrepareAndFinalize` abstract class exposes `prepare`, `prepare_no_receive` and `finalize` functions.
58
The `prepare` function is responsible for input activation Quantization and All2All Dispatch. If implemented, The `prepare_no_receive` is like `prepare` except it does not wait to receive results from other workers. Instead it returns a "receiver" callback that must be invoked to wait for the final results of worker. It is not required that this method is supported by all `FusedMoEPrepareAndFinalize` classes, but if it is available, it can be used to interleave work with the initial all to all communication, e.g. interleaving shared experts with fused experts. The `finalize` function is responsible for invoking the All2All Combine. Additionally the `finalize` function may or may not do the TopK weight application and reduction (Please refer to the TopKWeightAndReduce section)
59
60
-
60
+
61
62
### FusedMoEPermuteExpertsUnpermute
63
@@ -88,7 +88,7 @@ The core FusedMoE implementation performs a series of operations. It would be in
88
It is sometimes efficient to perform TopK weight application and Reduction inside the `FusedMoEPermuteExpertsUnpermute::apply()`. Find an example [here](https://github.com/vllm-project/vllm/pull/20228). We have a `TopKWeightAndReduce` abstract class to facilitate such implementations. Please refer to the TopKWeightAndReduce section.
89
`FusedMoEPermuteExpertsUnpermute::finalize_weight_and_reduce_impl()` returns the `TopKWeightAndReduce` object that the implementation wants the `FusedMoEPrepareAndFinalize::finalize()` to use.
90
91
-
91
+
92
93
### FusedMoEModularKernel
94