[Kernel] some optimizations for dense marlin and moe marlin (#16850)

Signed-off-by: Jinzhen Lin <linjinzhen@hotmail.com>

[Kernel] some optimizations for dense marlin and moe marlin (#16850)
Signed-off-by: Jinzhen Lin <linjinzhen@hotmail.com>
1d0c9d6b · Jinzhen Lin · GitHub · f62cad64 · 1d0c9d6b · 1d0c9d6b
Unverified Commit 1d0c9d6b authored May 06, 2025 by Jinzhen Lin Committed by GitHub May 05, 2025
6 changed files
--- a/vllm/model_executor/layers/quantization/fp8.py
+++ b/vllm/model_executor/layers/quantization/fp8.py
--- a/vllm/model_executor/layers/quantization/gptq_marlin.py
+++ b/vllm/model_executor/layers/quantization/gptq_marlin.py
--- a/vllm/model_executor/layers/quantization/kernels/mixed_precision/marlin.py
+++ b/vllm/model_executor/layers/quantization/kernels/mixed_precision/marlin.py
--- a/vllm/model_executor/layers/quantization/utils/marlin_utils.py
+++ b/vllm/model_executor/layers/quantization/utils/marlin_utils.py
--- a/vllm/model_executor/layers/quantization/utils/marlin_utils_fp8.py
+++ b/vllm/model_executor/layers/quantization/utils/marlin_utils_fp8.py
--- a/vllm/scalar_type.py
+++ b/vllm/scalar_type.py