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, 15:15 UTC
[ROCm][CI] Pinning timm lib version to fix ImportError in Multi-Modal Tests (Nemotron) (#31835)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Andreas Karatzas committed
Jan 6, 2026 at 17:23 UTC
e5d427e93af5861e22c2b7b3ce88af0028fc41e3
1 file changed
+2
requirements/rocm-test.txt
+2
@@ -86,3 +86,5 @@ arctic-inference == 0.1.1
86
open-clip-torch==2.32.0
87
# Required for isaac Multi-Modal generation test
88
perceptron==0.1.4
89
+# Required for the multi-modal models test
90
+timm==1.0.17