In this document, we list the installation requirements and installation scripts for different methods, as shown in the table below.
| Method | Python | CUDA | Key Dependencies | Install Command |
|---|---|---|---|---|
| 馃Л Navigation Video Generation | ||||
| MatrixGame2 | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| MatrixGame3 | 3.12 (recommended) | 12.1+ | PyTorch 2.10.0, transformers 4.57.3, FlashAttention (optional but recommended) | bash scripts/setup/matrix_game_3_install.shbash scripts/test_inference/test_nav_video_gen.sh matrix-game-3 |
| Hunyuan-GameCraft | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| HunyuanWorld-Voyager | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/lower_trans_install.sh |
| Hunyuan-WorldPlay | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| Astra | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| Yume | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| LingBot-World | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| LingBot-Video | 3.10+ | 13.0 | PyTorch 2.12.0 nightly (cu130), Diffusers 0.39.0, Transformers 5.8.1, SGLang 0.5.13 (optional) | bash scripts/setup/lingbot_video_install.shbash scripts/setup/lingbot_video_install.sh --sglang (SGLang) |
| Sana-WM | 3.10 | 12.1+ | PyTorch 2.5.1 | bash scripts/setup/sana_wm_install.sh |
| Infinite-World | 3.10 | 12.1 | PyTorch 2.5.1, flash-attn | bash scripts/setup/default_install.sh |
| Cosmos3 | 3.11 | 12.8+ | PyTorch, Diffusers main, Transformers, Cosmos Guardrail | bash scripts/setup/cosmos3_install.shbash scripts/test_inference/test_nav_video_gen.sh cosmos3 |
| GammaWorld | 3.10 | 12.8 | PyTorch, Transformer Engine, Gamma-World package | bash scripts/setup/gamma_world_install.shbash scripts/test_inference/test_nav_video_gen.sh gamma-world |
| Solaris | 3.10 | 12.1+ | JAX 0.6.2, Flax 0.10.6, PyTorch 2.8.0 | bash scripts/setup/solaris_install.shbash scripts/test_inference/test_nav_video_gen.sh solaris |
| MemFlow | 3.10 | 12.8 | PyTorch 2.8.0, Diffusers 0.31.0, FlashAttention | bash scripts/setup/memflow_install.shbash scripts/test_inference/test_nav_video_gen.sh memflow |
| 馃帹 3D Scene Generation | ||||
| FlashWorld | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/flash_world_install.sh |
| HunyuanWorld-Mirror | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/hunyuan_mirror_install.sh |
| Hunyuan-worldplay2 / HY-World-2.0 | 3.11.15 | 12.8 | PyTorch, WorldMirror 2.0, gsplat | bash scripts/setup/hunyuan_worldplay2_install.shbash scripts/test_inference/test_3d_scene_gen.sh hunyuan-worldplay2 |
| VGGT | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_3D_install.sh |
| Pi3 | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_3D_install.sh |
| Pi3X | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_3D_install.sh |
| LoGeR | 3.10 | 12.1 | PyTorch 2.6.0 | bash scripts/setup/loger_install.sh |
| InfiniteVGGT | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_3D_install.sh |
| FantasyWorld | 3.10 | 12.1+ | PyTorch 2.4.0+, Diffusers 0.31.0+, Transformers 4.49.0 | bash scripts/setup/fantasy_world_install.shbash scripts/test_inference/test_3d_scene_gen.sh fantasy-world |
| 馃 Vision Language Action | ||||
| 蟺0 | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| 蟺0.5 | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| GigaBrain-0 | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| Spirit v1.5 | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh
bash scripts/setup/libero_install.sh |
| Lingbot-va | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_lingbot_va.sh |
| Ctrl-World | 3.11 | 12.6+ | PyTorch 2.7.1, Diffusers 0.34.0, Transformers 4.48.1 | bash scripts/setup/ctrl_world_install.shbash scripts/test_inference/test_vla.sh ctrl-world |
| 馃帗 Multimodal Reasoning | ||||
| OmniVinci | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/omnivinci_install.sh |
| Qwen2.5-Omni | 3.10 | 12.1 | PyTorch 2.6.0 | bash scripts/setup/default_audio_install.sh |
| SpatialLadder | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| SpatialReasoner | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| 馃З Interactive Video | ||||
| Sora 2 | 3.10 | Only need API | ||
| Veo 3 | 3.10 | Only need API | ||
| Wan 2.5 | 3.10 | Only need API | ||
| Wan 2.2 | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| WoW | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| Cosmos-Predict 2.5 | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| Recammaster | 3.10 | 12.1 | PyTorch 2.5.1 | bash scripts/setup/default_install.sh |
| LongLive | 3.10 | 12.4 | PyTorch 2.5.0, flash-attn 2.7.4.post1 | See docs/longlive_openworldlib_migration.md |
| Rolling Forcing | 3.10 | 12.4 | PyTorch 2.5.1, flash-attn | See docs/rolling_forcing_openworldlib_migration.mdbash scripts/test_inference/test_nav_video_gen.sh rolling-forcing |
| 鈿欙笍 Simulation Environment | ||||
| AI2thor | 3.9 | bash scripts/setup/ai2thor_install.sh |
||
| 馃幍 Audio Generation | ||||
| MMAudio | 3.10 | 12.1 | PyTorch 2.6.0 | bash scripts/setup/audio_generation_default_install.sh |
| ThinkSound | 3.10 | 12.1 | PyTorch 2.6.0 | bash scripts/setup/audio_generation_default_install.sh |