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
[Doc] updating torch.compile doc link (#25989)
Signed-off-by: nadathurv <work.vnadathur@gmail.com> Signed-off-by: WorldExplored <srreyansh.sethi@gmail.com> Co-authored-by: Srreyansh Sethi <107075589+WorldExplored@users.noreply.github.com>
nadathurv committed
Oct 1, 2025 at 00:04 UTC
57b46d769eb7c6bfe03756da3071aa99779b1b99
1 file changed
+1
-1
docs/design/torch_compile.md
+1
-1
@@ -16,7 +16,7 @@ vLLM will take all the available factors into consideration, and decide a direct
16
17
The factors considered include:
18
19
-- All the related configs (see the `compute_hash` functions in the [config.py](gh-file:vllm/config.py))
19
+- All the related configs (see the `compute_hash` functions in their respective configs in the [config folder](gh-file:vllm/config))
20
- PyTorch configs (see the `compute_hash` functions in the [compiler_interface.py](gh-file:vllm/compilation/compiler_interface.py))
21
- The model's forward function and the relevant functions called by the forward function (see below)
22