frozenleaves/vllm
1
0
Fork 0
mirror of https://github.com/vllm-project/vllm.git synced 2025-10-20 14:53:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Doc] Fix invalid JSON in example args (#18527)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2025-05-22 15:11:46 +08:00
committed by GitHub
parent db5a29ba19
commit 23b67b37b2
1 changed files with 7 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
docs/source/design/v1/torch_compile.md
View File

@ -99,7 +99,9 @@ This time, Inductor compilation is completely bypassed, and we will load from di
The above example just uses Inductor to compile for a general shape (i.e. symbolic shape). We can also use Inductor to compile for some of the specific shapes, for example:
`vllm serve meta-llama/Llama-3.2-1B --compilation_config "{'compile_sizes': [1, 2, 4, 8]}"`
```
vllm serve meta-llama/Llama-3.2-1B --compilation_config '{"compile_sizes": [1, 2, 4, 8]}'
```
Then it will also compile a specific kernel just for batch size `1, 2, 4, 8`. At this time, all of the shapes in the computation graph are static and known, and we will turn on auto-tuning to tune for max performance. This can be slow when you run it for the first time, but the next time you run it, we can directly bypass the tuning and run the tuned kernel.
@ -134,12 +136,14 @@ The cudagraphs are captured and managed by the compiler backend, and replayed wh
By default, vLLM will try to determine a set of sizes to capture cudagraph. You can also override it using the config `cudagraph_capture_sizes`:
`vllm serve meta-llama/Llama-3.2-1B --compilation-config "{'cudagraph_capture_sizes': [1, 2, 4, 8]}"`
```
vllm serve meta-llama/Llama-3.2-1B --compilation-config '{"cudagraph_capture_sizes": [1, 2, 4, 8]}'
```
Then it will only capture cudagraph for the specified sizes. It can be useful to have fine-grained control over the cudagraph capture.
### Full Cudagraph capture
It is possible to include attention as part of the cudagraph if using an attention backend that is cudagraph compatible. This can improve performance in some cases such as decode speed for smaller models. Enable this using `--compilation-config "{'full_cuda_graph': True}"`
It is possible to include attention as part of the cudagraph if using an attention backend that is cudagraph compatible. This can improve performance in some cases such as decode speed for smaller models. Enable this using `--compilation-config '{"full_cuda_graph": true}'`.
Currently only FlashAttention 3 is compatible, and only when cascade attention is disabled.