Skip to content

fix(aiter): use tuned GEMM for unquantized linear with torchao compatibility guard #20889

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
michaelzhang-ai wants to merge 1 commit into from
Closed

fix(aiter): use tuned GEMM for unquantized linear with torchao compatibility guard #20889

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions python/sglang/srt/layers/quantization/unquant.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,7 @@
from aiter import ActivationType
from aiter.fused_moe import fused_moe
from aiter.ops.shuffle import shuffle_weight
from aiter.tuned_gemm import tgemm

if _is_npu:
from sglang.srt.hardware_backend.npu.utils import npu_format_cast
Expand Down Expand Up @@ -150,6 +151,9 @@ def apply(
output = output.view(x_shapes[0], x_shapes[1], -1)
return output

if _use_aiter and type(layer.weight.data) is torch.Tensor:
return tgemm.mm(x, layer.weight, bias, otype=x.dtype)

return F.linear(x, layer.weight, bias)


Expand Down
Loading