Any-to-Any
Transformers
Safetensors
English
Chinese
qwen3_vl
image-text-to-text
embodied-ai
vision-language-model
robotics
spatial-reasoning
multimodal
Instructions to use DeepCybo/PhysBrain1.5-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use DeepCybo/PhysBrain1.5-2B with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("DeepCybo/PhysBrain1.5-2B") model = AutoModelForMultimodalLM.from_pretrained("DeepCybo/PhysBrain1.5-2B", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update pipeline tag and add library name
#3
by nielsr HF Staff - opened
This PR updates the model card metadata to better reflect PhysBrain 1.5's capabilities and integration:
- Changes
pipeline_tagfromimage-text-to-texttoany-to-any, since the model unifies text, action-token generation, and future visual-state prediction in a single autoregressive framework. - Adds
library_name: transformers, enabling the appropriate Transformers inference snippet. The repository containsconfig.jsonwithQwen3VLForConditionalGeneration,qwen3_vl, and atransformers_version, which supports Transformers compatibility.
The existing model card content is preserved.
VLyb changed pull request status to merged