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 16, 17:00 UTC
[CI/Build] make pip install vllm work in macos (for import only) (#8118)
tomeras91 committed
Sep 3, 2024 at 22:37 UTC
c02638efb36007458b11710e0f7428cffac7cbe4
1 file changed
+2
-1
setup.py
+2
-1
@@ -362,7 +362,8 @@ def get_vllm_version() -> str:
362
version = find_version(get_path("vllm", "version.py"))
363
364
if _no_device():
365
- version += "+empty"
365
+ if envs.VLLM_TARGET_DEVICE == "empty":
366
+ version += "+empty"
367
elif _is_cuda():
368
cuda_version = str(get_nvcc_cuda_version())
369
if cuda_version != MAIN_CUDA_VERSION: