Skip to content

AML-3571: Add ORCA support for Blackwell RTX Pro 6000 #2

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

Open
anvdn wants to merge 2 commits into
base: v0.5.5.post3+hs
Choose a base branch
from
+9 −3
Open

AML-3571: Add ORCA support for Blackwell RTX Pro 6000 #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
1292127
disable deepgemm for blackwell GPU as it requires CUDA toolkit
anvdn Mar 1, 2026
e359d79
Bump SGLang version to 0.5.5.post3+hs7
anvdn Mar 1, 2026
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
2 changes: 1 addition & 1 deletion python/pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@ build-backend = "setuptools.build_meta"

[project]
name = "sglang"
version = "0.5.5.post3+hs5"
version = "0.5.5.post3+hs7"
description = "SGLang is a fast serving framework for large language models and vision language models."
readme = "README.md"
requires-python = ">=3.10"
Expand Down
10 changes: 8 additions & 2 deletions python/sglang/srt/layers/deep_gemm_wrapper/configurer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,8 +18,14 @@ def _compute_enable_deep_gemm():

return envs.SGLANG_ENABLE_JIT_DEEPGEMM.get()


ENABLE_JIT_DEEPGEMM = _compute_enable_deep_gemm()
# deepgemm requires CUDA toolkit and will error out on Blackwell GPU if missing.
# File "/.pyenv/versions/3.12.7/lib/python3.12/site-packages/deep_gemm/__init__.py", line 42, in _ensure_initialized
# torch.ops.deep_gemm.init(library_root, _find_cuda_home())
# ^^^^^^^^^^^^^^^^^
# File "/.pyenv/versions/3.12.7/lib/python3.12/site-packages/deep_gemm/__init__.py", line 30, in _find_cuda_home
# assert cuda_home is not None
# ^^^^^^^^^^^^^^^^^^^^^
ENABLE_JIT_DEEPGEMM = False

DEEPGEMM_BLACKWELL = ENABLE_JIT_DEEPGEMM and is_blackwell()
DEEPGEMM_SCALE_UE8M0 = DEEPGEMM_BLACKWELL