Skip to content

PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度

add llmc for cuda ut (#1145)

Yi Liu2aeffe78aa02
diff --git a/test/test_cpu/requirements.txt b/test/test_cpu/requirements.txt
index cfa43463c..219189829 100644
--- a/test/test_cpu/requirements.txt
+++ b/test/test_cpu/requirements.txt
@@ -6,6 +6,5 @@ torchvision
parameterized
pillow
numba
-#TODO: (yiliu30) replace it with the release version
llmcompressor @ git+https://github.com/vllm-project/llm-compressor.git@main
lm_eval
\ No newline at end of file
diff --git a/test/test_cuda/requirements.txt b/test/test_cuda/requirements.txt
index 6cdc59633..86c0435c7 100644
--- a/test/test_cuda/requirements.txt
+++ b/test/test_cuda/requirements.txt
@@ -13,4 +13,4 @@ pillow
torchvision
numba
vllm>=0.8.5.post1
-compressed-tensors
+llmcompressor @ git+https://github.com/vllm-project/llm-compressor.git@main
\ No newline at end of file