xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 10:46:08 +08:00
Code Issues Packages Projects Releases Wiki Activity

[ROCm] Temporarily remove GPTQ ROCm support (#2138)

Browse Source
This commit is contained in:
Woosuk Kwon 2023-12-15 17:13:58 -08:00 committed by GitHub
parent b81a6a6bb3
commit 2acd76f346
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
setup.py
View File

@ -219,13 +219,13 @@ vllm_extension_sources = [
"csrc/activation_kernels.cu", "csrc/activation_kernels.cu",
"csrc/layernorm_kernels.cu", "csrc/layernorm_kernels.cu",
"csrc/quantization/squeezellm/quant_cuda_kernel.cu", "csrc/quantization/squeezellm/quant_cuda_kernel.cu",
"csrc/quantization/gptq/q_gemm.cu",
"csrc/cuda_utils_kernels.cu", "csrc/cuda_utils_kernels.cu",
"csrc/pybind.cpp", "csrc/pybind.cpp",
] ]
if _is_cuda(): if _is_cuda():
vllm_extension_sources.append("csrc/quantization/awq/gemm_kernels.cu") vllm_extension_sources.append("csrc/quantization/awq/gemm_kernels.cu")
vllm_extension_sources.append("csrc/quantization/gptq/q_gemm.cu")
vllm_extension = CUDAExtension( vllm_extension = CUDAExtension(
name="vllm._C", name="vllm._C",

2
vllm/config.py
View File

@ -143,7 +143,7 @@ class ModelConfig:
def _verify_quantization(self) -> None: def _verify_quantization(self) -> None:
supported_quantization = ["awq", "gptq", "squeezellm"] supported_quantization = ["awq", "gptq", "squeezellm"]
rocm_not_supported_quantization = ["awq"] rocm_not_supported_quantization = ["awq", "gptq"]
if self.quantization is not None: if self.quantization is not None:
self.quantization = self.quantization.lower() self.quantization = self.quantization.lower()