Gemini Robotics On-Device brings AI to local robotic devices

Actually very close to one I'd say.

It's a "visual language action" VLA model "built on the foundations of Gemini 2.0".

As Gemini 2.0 has native language, audio and video support, I suspect it has been adapted to include native "action" data too, perhaps only on output fine-tuning rather than input/output at training stage (given its Gemini 2.0 foundation).

Natively multimodal LLM's are basically brains.

OpenVLA is basically a slightly modified, fine-tuned llama2. I found the launch/intro talk by lead author to be quite accessible: https://www.youtube.com/watch?v=-0s0v3q7mBk

> Natively multimodal LLM's are basically brains.

Absolutely not.

A more modern one, smolVLA is similar and uses a VLM but skips a few layers and uses an action adapter for outputs. Both are from HF and run on LeRobot.

https://arxiv.org/abs/2506.01844

Explanation by PhosphoAI: https://www.youtube.com/watch?v=00A6j02v450

Lol keep telling yourself that. It's not a human brain nor is it necessarily a very intelligent brain, but it is a brain nonetheless.

In the paper at the bottom of googles page, this VLA says it is built on the foundations of Gemini 2.0 (hence my quotations). They'd be using Gemini 2.0 rather than llama.

https://arxiv.org/pdf/2503.20020

Not a useful commentary. ANN and BNN are slightly correlated. That fact that you want to believe it is a brain tells a lot about you, but it doesn’t make a model a brain.

Only suggestion I have is “study more”.

They're not merely slightly correlated.

If it looks like a duck and quacks like a duck...

Just because it is alien to you, does not mean it is not a brain, please go look up the definition of the word.

And my comment is useful, a VLA implies it is processing it's input and output natively, something a brain does hence my comment.