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, 04:00 UTC
[ROCm][CI] Adding missing dependencies for Multi-modal models tests (#36177)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Andreas Karatzas committed
Mar 5, 2026 at 22:23 UTC
639680d220c9103cf47d63c5ff0ad3885426f487
1 file changed
+3
-1
requirements/rocm-test.txt
+3
-1
@@ -91,7 +91,7 @@ timm==1.0.17
91
# Required for plugins test
92
albumentations==1.4.6
93
# Pin transformers version
94
-transformers==4.57.3
94
+transformers==4.57.5
95
# Pin HF Hub version
96
huggingface-hub==0.36.2
97
# Pin Mistral Common
@@ -106,3 +106,5 @@ imagehash==4.3.2
106
bitsandbytes==0.49.2
107
# Examples (tensorizer) tests
108
tensorizer==2.10.1
109
+# Multi-modal models test (`allendou/FireRedASR2-LLM-vllm`)
110
+kaldi-native-fbank==1.22.3