Robotics
PyTorch
Cosmos
xperience10m_task_baseline_suite
embodied-ai
multimodal
xperience-10m
baseline
evaluation
qwen3-omni
Instructions to use cy0307/ropedia-xperience-10m-task-baselines with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Cosmos
How to use cy0307/ropedia-xperience-10m-task-baselines with Cosmos:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
| #!/usr/bin/env python3 | |
| """Runtime compatibility shims for local Qwen3-Omni Transformers builds.""" | |
| from __future__ import annotations | |
| def patch_qwen3_omni_config() -> None: | |
| """Patch a Transformers 5.0.0 Qwen3-Omni config init ordering bug. | |
| The local build reads `self.use_sliding_window` inside | |
| `Qwen3OmniMoeTalkerCodePredictorConfig.__init__` before `PreTrainedConfig` | |
| has copied unknown keyword arguments onto the instance. ModelScope's Instruct | |
| config includes `use_sliding_window`, but the attribute is not available yet. | |
| Setting it before the original initializer keeps the upstream behavior while | |
| allowing local `from_pretrained()` to proceed. | |
| """ | |
| try: | |
| from transformers.models.qwen3_omni_moe import configuration_qwen3_omni_moe as cfg | |
| except Exception: | |
| return | |
| cls = getattr(cfg, "Qwen3OmniMoeTalkerCodePredictorConfig", None) | |
| if cls is None or getattr(cls, "_ropedia_use_sliding_window_patch", False): | |
| return | |
| original_init = cls.__init__ | |
| def patched_init(self, *args, **kwargs): | |
| self.use_sliding_window = bool(kwargs.get("use_sliding_window", False)) | |
| original_init(self, *args, **kwargs) | |
| cls.__init__ = patched_init | |
| cls._ropedia_use_sliding_window_patch = True | |