GigaAI · robot AI model
GigaWorld-1
China · 2026
Sources retrieved: 9 November 2026 (A, B, C, D)
- Model type
- World modelSource and note: Model typeC huggingface.co/open-gigaai/Giga-World-1 · Retrieved 9 November 2026Note. Hugging Face model card: “a world model specially optimized for robot policy evaluation”.
- Inputs
- Camera LanguageSource and note: InputsA github.com/open-gigaai/giga-world-1 · Retrieved 9 November 2026Note. GitHub README, image-to-video mode: “Nano i2v: first frame + control video + text prompt”. Project page: “Historical frames, future noisy latents, and temporally aligned controls such as actions, depth, semantic maps, and captions are fused for embodied rollout generation”
- Access
- Open weights (download)Source and note: AccessC huggingface.co/open-gigaai/Giga-World-1 · Retrieved 9 November 2026Note. Hugging Face model card: “This repository contains checkpoints for” GigaWorld-1. The repository is listed as not gated.
- Base model
- WAN 2.1 1.3B FunControl; WAN 2.2 5B FunControlSource and note: Base modelC huggingface.co/open-gigaai/Giga-World-1 · Retrieved 9 November 2026Note. Hugging Face model card, base checkpoints before stage 1: “WAN 2.1 1.3B FunControl” and “WAN 2.2 5B FunControl”.
- Licence type
- Free, commercial useSource and note: Licence typeC huggingface.co/open-gigaai/Giga-World-1 · Retrieved 9 November 2026Note. “This model is released under the Apache License 2.0 unless otherwise specified.” GigaAI says GigaWorld-1 “provides open-source training, inference, data processing, checkpoint conversion, and LoRA merge workflows for robot world models.”
Specifications
7 of 14 stated 7 not stated
3 fields hidden 4 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
Released2 July 2026
Source and note: ReleasedD arxiv.org/abs/2607.02642 · Retrieved 9 November 2026Note. [A] “Submitted on 2 Jul 2026”
Input and output2/4 disclosed
Model typeWorld model
Source and note: Model typeC huggingface.co/open-gigaai/Giga-World-1 · Retrieved 9 November 2026Note. Hugging Face model card: “a world model specially optimized for robot policy evaluation”.
InputsCamera Language
Source and note: InputsA github.com/open-gigaai/giga-world-1 · Retrieved 9 November 2026Note. GitHub README, image-to-video mode: “Nano i2v: first frame + control video + text prompt”. Project page: “Historical frames, future noisy latents, and temporally aligned controls such as actions, depth, semantic maps, and captions are fused for embodied rollout generation”
What it controlsNot stated
Source and note: What it controlsB open-gigaai.github.io/giga-world-1/ · Retrieved 9 November 2026Note. The model generates video rollouts and does not itself command a robot: “each generated window is decoded and appended back into the rollout history.”
Degrees of freedom controlledNot stated
Size and speed0/2 disclosed
ParametersNot stated
Source and note: ParametersA github.com/open-gigaai/giga-world-1 · Retrieved 9 November 2026Note. GigaAI gives one size per variant and no total: “Supports both Nano (1.3B) and Pro (5B)”.
Control rateNot stated
Source and note: Control rateA github.com/open-gigaai/giga-world-1 · Retrieved 9 November 2026Note. The README gives the speed of the generated video, not a control rate: “Nano + Pro one-click scripts, 10 FPS, 33 s rollouts”.
Where it runs and what it is built on1/2 disclosed
On robot or cloudNot stated
Base modelWAN 2.1 1.3B FunControl; WAN 2.2 5B FunControl
Source and note: Base modelC huggingface.co/open-gigaai/Giga-World-1 · Retrieved 9 November 2026Note. Hugging Face model card, base checkpoints before stage 1: “WAN 2.1 1.3B FunControl” and “WAN 2.2 5B FunControl”.
Openness3/3 disclosed
AccessOpen weights (download)
Source and note: AccessC huggingface.co/open-gigaai/Giga-World-1 · Retrieved 9 November 2026Note. Hugging Face model card: “This repository contains checkpoints for” GigaWorld-1. The repository is listed as not gated.
Licence documentApache License 2.0
Source and note: Licence documentC huggingface.co/open-gigaai/Giga-World-1 · Retrieved 9 November 2026Note. “This model is released under the Apache License 2.0 unless otherwise specified.” Tag: “license: apache-2.0”.
Licence typeFree, commercial use
Source and note: Licence typeC huggingface.co/open-gigaai/Giga-World-1 · Retrieved 9 November 2026Note. “This model is released under the Apache License 2.0 unless otherwise specified.” GigaAI says GigaWorld-1 “provides open-source training, inference, data processing, checkpoint conversion, and LoRA merge workflows for robot world models.”
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
GigaAI describes GigaWorld-1 on Hugging Face as a world model specially optimized for robot policy evaluation.
GigaAI released two sizes: Nano (1.3B) and Pro (5B).
The stage-2 distilled checkpoints are not released yet.
Sources
One line per source, not per figure: a source has one retrieval date.
- Agithub.com/open-gigaai/giga-world-1Retrieved 9 November 2026
- Bopen-gigaai.github.io/giga-world-1/Main source. Retrieved 9 November 2026
- Chuggingface.co/open-gigaai/Giga-World-1Retrieved 9 November 2026
- Darxiv.org/abs/2607.02642Retrieved 9 November 2026
