minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 13 Packages Projects Releases Wiki Activity

[Kernel] Migrate scattered quantization kernels to sglang.kernels (RFC #29630, Phase 2.5, 1/7) (#30784)

Browse Source 
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Xiaoyu Zhang
2026-07-13 16:17:01 +08:00
committed by GitHub
co-authored by Claude Fable 5
parent 2225817424
commit 874fc07d9b
252 changed files with 294 additions and 214 deletions
Download Patch File Download Diff File Expand all files Collapse all files
benchmark/kernels/deepseek/benchmark_deepgemm_fp8_gemm.py
+1 -1
View File
@@ -12,7 +12,7 @@ from vllm.model_executor.layers.quantization.utils.fp8_utils import (
)
from sglang.benchmark.bench_utils import run_bench
from sglang.srt.layers.quantization.fp8_kernel import (
from sglang.kernels.ops.quantization.fp8_kernel import (
w8a8_block_fp8_matmul_deepgemm as w8a8_block_fp8_matmul,
)
benchmark/kernels/deepseek/benchmark_deepgemm_fp8_gemm_blackwell.py
+1 -1
View File
@@ -7,7 +7,7 @@ from deep_gemm import ceil_div
from flashinfer.gemm import gemm_fp8_nt_groupwise
from sglang.benchmark.bench_utils import run_bench
from sglang.srt.layers.quantization.fp8_kernel import (
from sglang.kernels.ops.quantization.fp8_kernel import (
sglang_per_token_group_quant_fp8,
w8a8_block_fp8_matmul_deepgemm,
)