biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Revert "[Performance] Move apply_w8a8_block_fp8_linear to an op class… (#25607)

Browse Source 
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
This commit is contained in:
Tyler Michael Smith
2025-09-25 04:05:21 -04:00
committed by GitHub
parent af4ee63e0e
commit 1260180c67
14 changed files with 205 additions and 346 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
benchmarks/cutlass_benchmarks/w8a8_benchmarks.py
View File

@@ -17,7 +17,7 @@ from weight_shapes import WEIGHT_SHAPES
from vllm import _custom_ops as ops
from vllm.model_executor.layers.quantization.utils.fp8_utils import (
w8a8_triton_block_scaled_mm,
w8a8_block_fp8_matmul,
)
from vllm.utils import FlexibleArgumentParser, cdiv
@@ -158,7 +158,7 @@ def bench_fp8(
"cutlass_fp8_fp8_fp16_scaled_mm_bias": lambda: ops.cutlass_scaled_mm(
a, b, scale_a, scale_b, torch.float16, bias.to(dtype=torch.float16)
),
"triton_fp8_fp8_fp16_scaled_mm_blockwise": lambda: w8a8_triton_block_scaled_mm(
"triton_fp8_fp8_fp16_scaled_mm_blockwise": lambda: w8a8_block_fp8_matmul(
a_cont, b.t(), block_scale_a, block_scale_b.t(), (128, 128)
),
"cutlass_fp8_fp8_fp16_scaled_mm_blockwise": lambda: ops.cutlass_scaled_mm(