DynamicWAM is an exact-time motion-conditioned world-action model for dynamic bimanual manipulation. It augments rendered history flow with numeric motion tokens that preserve displacement magnitude, elapsed simulator time, velocity, and acceleration.

Project page · Checkpoint · Apache-2.0 license

Method

DynamicWAM uses two complementary motion representations:

• History flow preserves where and in which direction the scene moved.
• Exact-time motion tokens preserve how far and how quickly it moved.

The released model conditions on four history-flow maps and four exact-time motion tokens; a 12-layer compact video expert and a 12-layer action expert exchange information through layer-wise joint attention.

Results — DOMINO Level 1

All 35 clean Level-1 tasks, 100 episodes per task, unseen instructions, native synchronous protocol with 16 committed joint-position actions per observation:

System	Motion input	Success	SR	MS
WAM reference	Current observation	795 / 3,500	22.71	38.32
History-Flow WAM	Four history-flow maps	953 / 3,500	27.23	41.62
DynamicWAM	History flow + exact-time motion tokens	1,337 / 3,500	38.20	53.16

These rows are complete systems with different training trajectories, not a matched one-variable ablation.

Installation

Linux x86_64 with an NVIDIA GPU, Python 3.10–3.12. The dependency graph is locked in uv.lock; FlashAttention installs separately because its wheel must match the local CUDA and GPU architecture:

git clone https://github.com/Autumn1337/DynamicWAM.git
cd DynamicWAM
uv sync --extra dev
uv pip install flash-attn==2.8.3.post1 --no-build-isolation

Checkpoint and external assets

Weights, simulator assets, and datasets stay outside Git; every revision and SHA-256 is pinned in manifests/.

hf download KhalilGao/DynamicWAM \
  --revision "925cbb7aef5033c924f809ae87479d39fe9f76ff" \
  --include "external/checkpoints/DynamicWAM_full.pt" \
  --include "configs/absolute_motion_v2.yaml" \
  --local-dir .
uv run python scripts/verify_checkpoints.py

uv run python scripts/prepare_external.py wan --purpose inference
uv run python scripts/bootstrap_domino.py
uv run python scripts/prepare_external.py robotwin-assets
uv run python scripts/prepare_external.py curobo-source

Evaluation

DOMINO evaluation runs in a separate Python 3.10 environment because the simulator and CuRobo require a different PyTorch/CUDA stack:

python3.10 -m venv external/robotwin
external/robotwin/bin/pip install -r environments/evaluation.txt
external/robotwin/bin/pip install -e . --no-deps

TORCH_CUDA_ARCH_LIST=9.0 \
  external/robotwin/bin/pip install -e external/curobo \
  --no-build-isolation --no-deps

external/robotwin/bin/python scripts/prepare_external.py domino-runtime
external/robotwin/bin/python scripts/eval_domino.py

Training

The canonical profile is configs/absolute_motion_v2.yaml. Training prompts are generated from the pinned DOMINO instruction generator (seen pool); evaluation generates unseen instructions at runtime. Fresh training also needs the History-Flow WAM initialization named under paths.base_checkpoint, which is not part of the minimal release.

uv run python scripts/prepare_external.py wan \
  --purpose language --purpose packing --purpose training

uv run python scripts/collect_domino.py
uv run python scripts/precompute_language.py
uv run python scripts/convert_domino.py
uv run python scripts/precompute_motion.py
uv run python scripts/pack_dataset.py
scripts/train_all.sh

scripts/train.py exposes the three stages individually: video-expert distillation, action-expert training, and joint fine-tuning.

License

Project-owned code is released under the Apache License 2.0. Vendored WAN-derived files and the DOMINO patch retain their upstream licenses. CuRobo is an external dependency under NVIDIA's non-commercial research/evaluation terms; the project license does not override that restriction.