GR-2
No estimate
No hardware requirements for this model
The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.
On record
Full specification
Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.
Origin
Who built this model, where, and when it was published.
- Organisation
- ByteDance
- Organisation type
- Industry
- Country
- China
- Published
- 8 October 2024
- Authors
- Hang Li, Yifeng Li, Yuxiao Liu, Hongtao Wu, Jiafeng Xu, Yichu Yang, Hanbo Zhang, Minzhao Zhu
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Robotics
- Task
- Video, Action recognition, Video generation, Instruction interpretation, Robotic manipulation
Size
How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.
- Parameters
- 230M
- Training data
- tokens
the default GR-2 model contains 230M parameters, of which 95M are trainable
This large-scale pre-training, involving 38 million video clips and over 50 billion tokens
Availability
Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.
- Weights
- Closed — provider access only
- Model access
- Unreleased
- Training code
- Unreleased
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Why it is tracked
- SOTA improvement
- Record confidence
- Confident
- Citations
- 224
Top perfoming model on CALVIN benchmark.
Sources
Where this record came from and when it was last checked.
- Reference
- GR-2: A Generative Video-Language-Action Model with Web-Scale Knowledge for Robot Manipulation
- Last updated
- 25 May 2026
What the numbers mean
Where it came from
GR-2 was published by ByteDance, in the country recorded as China, during October 2024. The category the publisher falls under is industry.
It works in the domain of Robotics, and is recorded as performing the task of video, Action recognition, Video generation, Instruction interpretation, Robotic manipulation.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
Training and provenance
It is tracked in the underlying dataset for one reason in particular: sOTA improvement.
Answers
GR-2 — common questions
GR-2— what GPU do I need to run it?
None. This is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.
GR-2— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
GR-2— how many parameters does it have?
It has a parameter count of 230M. the default GR-2 model contains 230M parameters, of which 95M are trainable. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.
GR-2— who created it?
It was published by ByteDance, based in China, an organisation categorised as industry.
GR-2— when was it released?
It was published in October 2024. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
GR-2— what is it used for?
It works in the domain of Robotics, and is recorded as handling the task of video, Action recognition, Video generation, Instruction interpretation, Robotic manipulation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
The other direction
Looking at it from the other side?
This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.