Northeastern University · robot AI model
PhyWorld
United States · 2026
Sources retrieved: 19 November 2026 (A, B)
- Model type
- World modelSource and note: Model typeA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card: “PhyWorld is a physics-faithful video world model post-trained from”
- Inputs
- Camera LanguageSource and note: InputsA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card: “Video-to-video scene continuation”; “English prompts were used in the reported evaluation”.
- Access
- Open weights (download)Source and note: AccessA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card: “hf download NU-World-Model-Embodied-AI/phyworld”
- Base model
- Wan2.2-I2V-A14BSource and note: Base modelA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card: “Wan2.2-I2V-A14B”
- Licence type
- Free, commercial useSource and note: Licence typeA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card metadata: “license: apache-2.0”

Specifications
7 of 14 stated 7 not stated
4 fields hidden 3 not-stated fields with a note stay ·
How to read the figuresShow guideClose
- not statedNo source
- 0Reported as zero
- ASource
Select a source marker for source details and notes.
Release1/1 disclosed
Released19 May 2026
Source and note: ReleasedB arxiv.org/abs/2605.19242 · Retrieved 19 November 2026Note. arXiv abstract page: “Submitted on 19 May 2026”
Input and output2/4 disclosed
Model typeWorld model
Source and note: Model typeA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card: “PhyWorld is a physics-faithful video world model post-trained from”
InputsCamera Language
Source and note: InputsA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card: “Video-to-video scene continuation”; “English prompts were used in the reported evaluation”.
What it controlsNot stated
Source and note: What it controlsA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Primary task is video-to-video scene continuation rather than direct robot control: “Video-to-video scene continuation”
Degrees of freedom controlledNot stated
Size and speed0/2 disclosed
ParametersNot stated
Source and note: ParametersA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Post-trained from “Wan2.2-I2V-A14B”.
Control rateNot stated
Where it runs and what it is built on1/2 disclosed
On robot or cloudNot stated
Base modelWan2.2-I2V-A14B
Source and note: Base modelA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card: “Wan2.2-I2V-A14B”
Openness3/3 disclosed
AccessOpen weights (download)
Source and note: AccessA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card: “hf download NU-World-Model-Embodied-AI/phyworld”
Licence documentApache-2.0
Source and note: Licence documentA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card metadata: “license: apache-2.0”
Licence typeFree, commercial use
Source and note: Licence typeA huggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.md · Retrieved 19 November 2026Note. Model card metadata: “license: apache-2.0”
The developer’s own claims0/2 disclosed
Learns new tasksNot stated
Works across robotsNot stated
Source and note: Works across robotsYes only if the developer itself says the model runs on different robot models, for example robots from two manufacturers or two of one manufacturer’s models. Several units of the same robot do not count.
Notes
Developed by researchers at Northeastern University (Boston, MA), University of Georgia, Tulane University, and EmbodyX. Submitted to arXiv on 19 May 2026.
The Hugging Face repository provides downloadable checkpoints, though the public code repository does not yet include an end-to-end inference command.
Sources
One line per source, not per figure: a source has one retrieval date.
- Ahuggingface.co/NU-World-Model-Embodied-AI/phyworld/raw/main/README.mdMain source. Retrieved 19 November 2026
- Barxiv.org/abs/2605.19242Retrieved 19 November 2026

