GigaAI · robot AI model
GigaWorld-Policy-0.5
China · 2026
Sources retrieved: 19 November 2026 (A, B, C)
- Model type
- World action modelSource and note: Model typeB github.com/open-gigaai/giga-world-policy · Retrieved 19 November 2026Note. Repository description: “A World Action Model enabling local real-time deployment with 85ms low latency.”
- Inputs
- Camera Language Robot stateSource and note: InputsA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Model summary: “The MoT design keeps a **visual expert** stream (reference + future latents) and an **action expert** stream (state + action), with multi-modal self-attention across both.” “| Text dim (T5) | 4096 |”
- Access
- Open weights (download)Source and note: AccessA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Hugging Face download command: “huggingface-cli download open-gigaai/Giga-World-Policy-0.5”
- Base model
- Wan2.2Source and note: Base modelA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Model card: “runtime also needs the Wan2.2 VAE / scheduler from the base Diffusers checkpoint.”
- Licence type
- Free, commercial useSource and note: Licence typeA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Model card header: “license: apache-2.0”

Specifications
8 of 14 stated 6 not stated
4 fields hidden 2 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
Released15 July 2026
Source and note: ReleasedC arxiv.org/abs/2607.13960 · Retrieved 19 November 2026Note. arXiv submission date: “Submitted on 15 Jul 2026”
Input and output2/4 disclosed
Model typeWorld action model
Source and note: Model typeB github.com/open-gigaai/giga-world-policy · Retrieved 19 November 2026Note. Repository description: “A World Action Model enabling local real-time deployment with 85ms low latency.”
InputsCamera Language Robot state
Source and note: InputsA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Model summary: “The MoT design keeps a **visual expert** stream (reference + future latents) and an **action expert** stream (state + action), with multi-modal self-attention across both.” “| Text dim (T5) | 4096 |”
What it controlsNot stated
Degrees of freedom controlledNot stated
Size and speed0/2 disclosed
ParametersNot stated
Control rateNot stated
Source and note: Control rateB github.com/open-gigaai/giga-world-policy · Retrieved 19 November 2026Note. States inference latency rather than Hz: “A World Action Model enabling local real-time deployment with 85ms low latency.”
Where it runs and what it is built on1/2 disclosed
On robot or cloudNot stated
Base modelWan2.2
Source and note: Base modelA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Model card: “runtime also needs the Wan2.2 VAE / scheduler from the base Diffusers checkpoint.”
Openness3/3 disclosed
AccessOpen weights (download)
Source and note: AccessA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Hugging Face download command: “huggingface-cli download open-gigaai/Giga-World-Policy-0.5”
Licence documentApache-2.0
Source and note: Licence documentA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Model card header: “license: apache-2.0”
Licence typeFree, commercial use
Source and note: Licence typeA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Model card header: “license: apache-2.0”
The developer’s own claims1/2 disclosed
Learns new tasksNeeds retraining
Source and note: Learns new tasksB github.com/open-gigaai/giga-world-policy · Retrieved 19 November 2026Note. Repository training guide: “Set the dataset, normalization, WAN base model, and initialization checkpoint paths before launching post-training.”
Works across robotsNot stated
Source and note: Works across robotsA huggingface.co/open-gigaai/Giga-World-Policy-0.5 · Retrieved 19 November 2026Note. Model summary lists “| Embodiments | 2 |”, without naming multiple robot models.Yes 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
GigaAI describes GigaWorld-Policy-0.5: “A World Action Model enabling local real-time deployment with 85ms low latency.”
Sources
One line per source, not per figure: a source has one retrieval date.
- Ahuggingface.co/open-gigaai/Giga-World-Policy-0.5Main source. Retrieved 19 November 2026
- Bgithub.com/open-gigaai/giga-world-policyRetrieved 19 November 2026
- Carxiv.org/abs/2607.13960Retrieved 19 November 2026




