GigaAI · robot AI model
GigaBrain-0.5M*
China · 2026
Sources retrieved: 13 November 2026 (A, B, C)
- Model type
- Vision-language-action (VLA)Source and note: Model typeC gigabrain05m.github.io · Retrieved 13 November 2026Note. Developer wording: “GigaBrain-0.5M* is a world model-conditioned VLA trained via world model-based reinforcement learning.” Project page abstract: “we propose GigaBrain-0.5M*, a VLA model trained via world model-based reinforcement learning”.
- What it controls
- Arms and handsSource and note: What it controlsA arxiv.org/html/2602.12099 · Retrieved 13 November 2026Note. Technical report, describing the GigaBrain-0.5 base: the model is “designed to map visual observations and language instructions to action sequences for bi-manual robots”. For GigaBrain-0.5M*, “The second phase initializes the policy from the pretrained” GigaBrain-0.5 checkpoint.
- Inputs
- Camera LanguageSource and note: InputsA arxiv.org/html/2602.12099 · Retrieved 13 November 2026Note. Technical report, describing the GigaBrain-0.5 base: the model is “designed to map visual observations and language instructions to action sequences for bi-manual robots”. For GigaBrain-0.5M*, “The second phase initializes the policy from the pretrained” GigaBrain-0.5 checkpoint.
- Base model
- GigaBrain-0.5Source and note: Base modelC gigabrain05m.github.io · Retrieved 13 November 2026Note. Project page abstract: “Built upon GigaBrain-0.5, which is pre-trained on over 10,000 hours of robotic manipulation data”.
Specifications
5 of 14 stated 9 not stated
8 fields hidden 1 not-stated field with a note stays ·
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
Released12 February 2026
Source and note: ReleasedB arxiv.org/abs/2602.12099 · Retrieved 13 November 2026Note. arXiv listing: “Submitted on 12 Feb 2026”.
Input and output3/4 disclosed
Model typeVision-language-action (VLA)
Source and note: Model typeC gigabrain05m.github.io · Retrieved 13 November 2026Note. Developer wording: “GigaBrain-0.5M* is a world model-conditioned VLA trained via world model-based reinforcement learning.” Project page abstract: “we propose GigaBrain-0.5M*, a VLA model trained via world model-based reinforcement learning”.
InputsCamera Language
Source and note: InputsA arxiv.org/html/2602.12099 · Retrieved 13 November 2026Note. Technical report, describing the GigaBrain-0.5 base: the model is “designed to map visual observations and language instructions to action sequences for bi-manual robots”. For GigaBrain-0.5M*, “The second phase initializes the policy from the pretrained” GigaBrain-0.5 checkpoint.
What it controlsArms and hands
Source and note: What it controlsA arxiv.org/html/2602.12099 · Retrieved 13 November 2026Note. Technical report, describing the GigaBrain-0.5 base: the model is “designed to map visual observations and language instructions to action sequences for bi-manual robots”. For GigaBrain-0.5M*, “The second phase initializes the policy from the pretrained” GigaBrain-0.5 checkpoint.
Degrees of freedom controlledNot stated
Size and speed0/2 disclosed
ParametersNot stated
Control rateNot stated
Where it runs and what it is built on1/2 disclosed
On robot or cloudNot stated
Base modelGigaBrain-0.5
Source and note: Base modelC gigabrain05m.github.io · Retrieved 13 November 2026Note. Project page abstract: “Built upon GigaBrain-0.5, which is pre-trained on over 10,000 hours of robotic manipulation data”.
Openness0/3 disclosed
AccessNot stated
Licence documentNot stated
Licence typeNot stated
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
The project page and the paper present the model; the project page links its Models button to the developer's Hugging Face organisation page, which lists no GigaBrain-0.5 or 0.5M* repository.
Sources
One line per source, not per figure: a source has one retrieval date.
- Aarxiv.org/html/2602.12099Retrieved 13 November 2026
- Barxiv.org/abs/2602.12099Retrieved 13 November 2026
- Cgigabrain05m.github.ioMain source. Retrieved 13 November 2026
