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 7, 06:30 UTC
docs: add version requirement note for --profiler-config flag (#32454)
Signed-off-by: abhishkh <mathukiya.a@northeastern.edu>
Abhishek Mathukiya committed
Mar 4, 2026 at 13:13 UTC
f3dc292e9f2cad55f914b7a7ed73e1969174ad77
1 file changed
+6
-2
docs/contributing/profiling.md
+6
-2
@@ -5,8 +5,12 @@
5
6
## Profile with PyTorch Profiler
7
8
-We support tracing vLLM workers using the `torch.profiler` module. You can enable the torch profiler by setting `--profiler-config`
9
-when launching the server, and setting the entries `profiler` to `'torch'` and `torch_profiler_dir` to the directory where you want to save the traces. Additionally, you can control the profiling content by specifying the following additional arguments in the config:
8
+We support tracing vLLM workers using different profilers. You can enable profiling by setting the `--profiler-config` flag when launching the server.
9
+
10
+!!! note
11
+ The `--profiler-config` flag is available in vLLM v0.13.0 and later. If you are using an earlier version, please upgrade to use this feature.
12
+
13
+To use the `torch.profiler` module, set the `profiler` entry to `'torch'` and `torch_profiler_dir` to the directory where you want to save the traces. Additionally, you can control the profiling content by specifying the following additional arguments in the config:
14
15
- `torch_profiler_record_shapes` to enable recording Tensor Shapes, off by default
16
- `torch_profiler_with_memory` to record memory, off by default