Skip to content

clarify deployment requirements for gr00t_n1_5 #1176

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
Teeeio wants to merge 5 commits into
base: main
Choose a base branch
from
+34 −1
Open

clarify deployment requirements for gr00t_n1_5 #1176

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
2391fcf
Update GR00T N1.5 README dependencies
Teeeio Mar 29, 2026
9ac403a
Add FFmpeg install instructions to GR00T README
Teeeio Mar 30, 2026
6bb804c
Pin peft version in GR00T README
Teeeio Mar 30, 2026
9b11693
Update GR00T requirements for single-command install
Teeeio Mar 31, 2026
1e0dbff
Merge branch 'main' into pr-1151
Teeeio Apr 2, 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
22 changes: 21 additions & 1 deletion examples/gr00t_n1_5/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,8 +24,28 @@ Install FlagScale and robotics dependencies:

```sh
cd FlagScale/
# replace "[cuda-train]" with "[ascend-train]" on Huawei Ascend, or "[musa-train]" on Moore Threads MUSA
# For NVIDIA CUDA training
pip install ".[cuda-train]" --verbose

# For NVIDIA CUDA serving
pip install ".[cuda-serve]" --verbose

# Replace [cuda-train] in the training command above with [ascend-train]
# on Huawei Ascend, or [musa-train] on Moore Threads MUSA.
```

GR00T N1.5 requires FFmpeg runtime libraries for `torchcodec`.
Install FFmpeg before running training or dataset preprocessing.

Install FFmpeg using one of the following options:

```sh
# Using conda
conda install -c conda-forge ffmpeg

# Or on Ubuntu/Debian
sudo apt-get update
sudo apt-get install -y ffmpeg
```

Install additional dependencies for downloading models/datasets:
Expand Down
3 changes: 3 additions & 0 deletions requirements/ascend/serve.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,9 @@

-r ./base.txt

# for robotics
peft==0.18.1

# FlagTree for Ascend (extra index so PyPI remains available for build deps)
# [--extra-index-url https://resource.flagos.net/repository/flagos-pypi-hosted/simple]
flagtree==0.4.0+ascend3.2
Expand Down
4 changes: 4 additions & 0 deletions requirements/cuda/serve.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,10 @@
# serve-specific dependencies

-r ./base.txt

# for robotics
peft==0.18.1

vllm @ https://resource.flagos.net/repository/flagos-pypi-hosted/packages/vllm/0.13.0%2Bfl.0.1.cu128.g72506c983/vllm-0.13.0%2Bfl.0.1.cu128.g72506c983-cp312-cp312-linux_x86_64.whl

# support 0.5b_multiple_instance ci test
Expand Down
6 changes: 6 additions & 0 deletions requirements/cuda/train.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,4 +4,10 @@
sentencepiece==0.2.1
transformers==4.57.6
tiktoken==0.12.0

# for robotics
peft==0.18.1
torchcodec==0.8.1
flash-attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.8.1/flash_attn-2.8.1+cu12torch2.9cxx11abiTRUE-cp312-cp312-linux_x86_64.whl

megatron_core @ https://resource.flagos.net/repository/flagos-pypi-hosted/packages/megatron-core/0.16.0rc0/megatron_core-0.16.0rc0-cp312-cp312-linux_x86_64.whl
Loading