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
Update to transformers==4.51.1 (#16257)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Harry Mellor committed
Apr 8, 2025 at 14:53 UTC
90cb44eb0205b6beca6311411113555a839e9d01
3 files changed
+3
-3
requirements/common.txt
+1
-1
@@ -6,7 +6,7 @@ requests >= 2.26.0
6
tqdm
7
blake3
8
py-cpuinfo
9
-transformers >= 4.51.0
9
+transformers >= 4.51.1
10
huggingface-hub[hf_xet] >= 0.30.0 # Required for Xet downloads.
11
tokenizers >= 0.19.1 # Required for Llama 3.
12
protobuf # Required by LlamaTokenizer.
requirements/test.in
+1
-1
@@ -31,7 +31,7 @@ mistral_common[opencv] >= 1.5.4 # required for pixtral test
31
opencv-python-headless >= 4.11.0 # required for video test
32
datamodel_code_generator # required for minicpm3 test
33
lm-eval[api]==0.4.8 # required for model evaluation test
34
-transformers==4.51.0
34
+transformers==4.51.1
35
huggingface-hub[hf_xet]>=0.30.0 # Required for Xet downloads.
36
# quantization
37
bitsandbytes>=0.45.3
requirements/test.txt
+1
-1
@@ -648,7 +648,7 @@ tqdm==4.66.6
648
# transformers
649
tqdm-multiprocess==0.0.11
650
# via lm-eval
651
-transformers==4.51.0
651
+transformers==4.51.1
652
# via
653
# -r requirements/test.in
654
# genai-perf