Kimodo Models
Several variations of Kimodo-v1 are available trained on various skeletons and datasets. All models support text-to-motion and kinematic controls.
| Model | Skeleton | Training Data | Release Date | License | Download |
|---|---|---|---|---|---|
| Kimodo-SOMA-RP-v1 | SOMA | Rigplay 1 (700hrs) | March 16, 2026 | NVIDIA Open Model | HF |
| Kimodo-G1-RP-v1 | G1 | Rigplay 1 (700hrs) | March 16, 2026 | NVIDIA Open Model | HF |
| Kimodo-SOMA-SEED-v1 | SOMA | BONES-SEED (288hrs) | March 16, 2026 | NVIDIA Open Model | HF |
| Kimodo-G1-SEED-v1 | G1 | BONES-SEED (288hrs) | March 16, 2026 | NVIDIA Open Model | HF |
| Kimodo-SMPLX-RP-v1 | SMPL-X | Rigplay 1 (700hrs) | March 16, 2026 | NVIDIA R&D Model | HF |
Which Model Should I Use?
For best quality motion generation, use models trained on the full Rigplay 1 dataset (700 hours).
For benchmarking and comparison, use BONES-SEED models (288 hours of publicly available data).
For humanoid robotics, use G1 skeleton models with MuJoCo-compatible output.
For compatibility with existing motion pipelines, use the SMPL-X model with AMASS format output.