Enable GPT-OSS TinyGEMM on CUDA 13 (#31649)
This commit is contained in:
@@ -72,7 +72,6 @@ from sglang.srt.runtime_context import get_forward, get_parallel, get_server_arg
|
||||
from sglang.srt.utils import (
|
||||
LazyValue,
|
||||
add_prefix,
|
||||
get_cuda_version,
|
||||
is_blackwell_supported,
|
||||
is_cpu,
|
||||
is_cuda,
|
||||
@@ -94,7 +93,7 @@ _is_tinygemm_supported = (
|
||||
and (is_sm90_supported() or is_blackwell_supported())
|
||||
)
|
||||
|
||||
if _is_tinygemm_supported and get_cuda_version()[0] < 13:
|
||||
if _is_tinygemm_supported:
|
||||
try:
|
||||
from flashinfer.gemm import tinygemm_bf16
|
||||
except ImportError:
|
||||
|
||||
Reference in New Issue
Block a user