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 15, 14:45 UTC
Fix triton install condition on CPU (#16600)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Harry Mellor committed
Apr 14, 2025 at 18:06 UTC
9883a1885970f53f88fd962c85dd406ade0966cd
1 file changed
+2
-2
requirements/cpu.txt
+2
-2
@@ -12,9 +12,9 @@ torchaudio; platform_machine != "ppc64le" and platform_machine != "s390x"
12
torchaudio==2.6.0; platform_machine == "ppc64le"
13
14
# required for the image processor of phi3v, this must be updated alongside torch
15
-torchvision; platform_machine != "ppc64le" and platform_machine != "s390x"
15
+torchvision; platform_machine != "ppc64le" and platform_machine != "s390x"
16
torchvision==0.21.0; platform_machine == "ppc64le"
17
datasets # for benchmark scripts
18
19
# cpu cannot use triton 3.3.0
20
-triton==3.2.0; platform_machine != "ppc64le"
20
+triton==3.2.0; platform_machine == "x86_64"