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 10, 06:30 UTC
[Doc] Update installation from source command (#32239)
Signed-off-by: esmeetu <jasonailu87@gmail.com>
Roy Wang committed
Jan 13, 2026 at 15:10 UTC
44c34f22d99a47e92715a74a2ca4ffbe919b2cbf
1 file changed
+8
-2
docs/contributing/README.md
+8
-2
@@ -43,10 +43,16 @@ If you are only developing vLLM's Python code, install vLLM using:
43
VLLM_USE_PRECOMPILED=1 uv pip install -e .
44
```
45
46
-If you are developing vLLM's Python and CUDA/C++ code, install vLLM using:
46
+If you are developing vLLM's Python and CUDA/C++ code, install Pytorch first:
47
48
```bash
49
-uv pip install -e .
49
+uv pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu129
50
+```
51
+
52
+then install vLLM using:
53
+
54
+```bash
55
+uv pip install -e . --no-build-isolation
56
```
57
58
For more details about installing from source and installing for other hardware, check out the [installation instructions](../getting_started/installation/README.md) for your hardware and head to the "Build wheel from source" section.