Skip to content

feat: Docker Blackwell / Grace Blackwell container support #2134

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
sahgerlad wants to merge 7 commits into
base: main
Choose a base branch
from
Open

feat: Docker Blackwell / Grace Blackwell container support #2134

Changes from 1 commit
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
3cd7284
feat(docker): Blackwell / Grace Blackwell container support
sahgerlad Mar 21, 2026
2c32624
Harden the find and guarding per-directory
sahgerlad Mar 21, 2026
11afc66
Merge branch 'main' into feat/docker-blackwell-support
sahgerlad Mar 23, 2026
ffde090
Update CUDA_ARCH_LIST error message
sahgerlad Mar 23, 2026
160da9f
Merge branch 'main' into feat/docker-blackwell-support
sahgerlad Mar 24, 2026
0ca4fcd
Merge branch 'main' into feat/docker-blackwell-support
terrykong Mar 31, 2026
bebce4c
Merge branch 'main' into feat/docker-blackwell-support
terrykong Apr 18, 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
6 changes: 4 additions & 2 deletions docker/Dockerfile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,6 +72,7 @@ RUN GITHUB_ARTIFACTORY=github.com \
ARG UV_VERSION=0.9.7
ARG PYTHON_VERSION=3.12
ENV PATH="/root/.local/bin:$PATH"
ENV TRITON_PTXAS_PATH=/usr/local/cuda/bin/ptxas
RUN curl -LsSf https://astral.sh/uv/${UV_VERSION}/install.sh | sh && \
uv python install ${PYTHON_VERSION}

Expand Down Expand Up @@ -109,8 +110,8 @@ ARG SKIP_SGLANG_BUILD
ENV UV_PROJECT_ENVIRONMENT=/opt/nemo_rl_venv
ENV UV_LINK_MODE=copy

# Ensure DeepEP is built for H100 and B200 (also mcore inference unified memory API now invokes a torch API that requires these to be set)
ENV TORCH_CUDA_ARCH_LIST="9.0 10.0"
# Ensure DeepEP is built for hopper and (grace)blackwell (also mcore inference unified memory API now invokes a torch API that requires these to be set)
ENV TORCH_CUDA_ARCH_LIST="9.0 10.0 10.3"
Comment thread
terrykong marked this conversation as resolved.

# First copy only the dependency files
COPY --from=nemo-rl pyproject.toml uv.lock ./
Expand Down Expand Up @@ -197,6 +198,7 @@ else
UV_LINK_MODE=symlink uv run nemo_rl/utils/prefetch_venvs.py
fi
EOF
RUN find /opt/ray_venvs /opt/nemo_rl_venv -name "ptxas-blackwell" -exec ln -sf /usr/local/cuda/bin/ptxas {} \;

# Generate container fingerprint for frozen environment support
# Store outside /opt/nemo-rl to avoid being overwritten by user mounts
Expand Down
Loading