[Refactor] Rename gptq_marlin to marlin to match MoE (#32952)

Signed-off-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
Michael Goin
2026-01-23 16:48:12 -05:00
committed by GitHub
parent 6cc6d92be5
commit 4561f13985
24 changed files with 40 additions and 40 deletions
+2 -2
View File
@@ -303,9 +303,9 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
ops.def("permute_cols(Tensor A, Tensor perm) -> Tensor");
ops.impl("permute_cols", torch::kCUDA, &permute_cols);
// gptq_marlin Optimized Quantized GEMM for GPTQ.
// Marlin Optimized Quantized GEMM (supports GPTQ, AWQ, FP8, NVFP4, MXFP4).
ops.def(
"gptq_marlin_gemm(Tensor a, Tensor? c_or_none, Tensor b_q_weight, "
"marlin_gemm(Tensor a, Tensor? c_or_none, Tensor b_q_weight, "
"Tensor? b_bias_or_none,Tensor b_scales, "
"Tensor? a_scales, Tensor? global_scale, Tensor? b_zeros_or_none, "
"Tensor? "