[Kernel][Performance] Add FlashInfer cutedsl NVFP4 GEMM backend (#42235)
Signed-off-by: Mohammad Miadh Angkad <176301910+mmangkad@users.noreply.github.com>
This commit is contained in:
@@ -141,6 +141,7 @@ LinearBackend = Literal[
|
||||
"auto",
|
||||
"cutlass",
|
||||
"flashinfer_cutlass",
|
||||
"flashinfer_cutedsl",
|
||||
"flashinfer_trtllm",
|
||||
"flashinfer_cudnn",
|
||||
"flashinfer_b12x",
|
||||
@@ -198,6 +199,7 @@ class KernelConfig:
|
||||
- "auto": Automatically select the best backend based on model and hardware
|
||||
- "cutlass": Use CUTLASS-based kernels
|
||||
- "flashinfer_cutlass": Use FlashInfer with CUTLASS kernels
|
||||
- "flashinfer_cutedsl": Use FlashInfer with CuteDSL kernels
|
||||
- "flashinfer_trtllm": Use FlashInfer with TensorRT-LLM kernels
|
||||
- "flashinfer_cudnn": Use FlashInfer with cuDNN kernels
|
||||
- "flashinfer_b12x": Use FlashInfer b12x CuteDSL NVFP4 GEMM (SM120+)
|
||||
|
||||
Reference in New Issue
Block a user