AWQ: Up to 2.66x higher throughput (#2566)
This commit is contained in:
@@ -51,6 +51,7 @@ PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
|
||||
#ifndef USE_ROCM
|
||||
// Quantization ops
|
||||
ops.def("awq_gemm", &awq_gemm, "Quantized GEMM for AWQ");
|
||||
ops.def("awq_dequantize", &awq_dequantize, "Dequantization for AWQ");
|
||||
#endif
|
||||
ops.def("gptq_gemm", &gptq_gemm, "Quantized GEMM for GPTQ");
|
||||
ops.def("gptq_shuffle", &gptq_shuffle, "Post processing for GPTQ");
|
||||
|
||||
Reference in New Issue
Block a user