We are especially excited about how Ο0.7 seems to exhibit emergent compositional generalization: it can put together skills it learned in new ways based on the prompt, for example to figure out how to use an air fryer to cook a sweet potato.
I'm extremely excited to announce that we've successfully inferenced Ο0.5 on our excavator!
We've collected a massive corpus of real-world data with natural language labels from operators in the industry and are using it to create some really cool policies. Here's our first demo of it successfully completing a task with just 200 trajectories. More on the way :)
Read our blog post: https://t.co/Y5poo1vetf
@physical_int
Weβve developed a memory system for our models that provides both short-term visual memory and long-term semantic memory.
Our approach allows us to train robots to perform long and complex tasks, like cleaning up a kitchen or preparing a grilled cheese sandwich from scratch π
new year, new job? we're hiring across many roles: https://t.co/m5w7u7OZ30 -- we're *just* starting to grow out the business side and posted our first business role. this role will be responsible for helping to scale the company and our deployments/partnerships/compute.
opening doors, using keys, washing dishes are still hard problems for robots. the Robot Olympics were a reminder of why and a delightful way to see progress show up in such ordinary ways :)
https://t.co/fc4ZBlnL0B
We got our robots to wash pans, clean windows, make peanut butter sandwiches, and more!
Fine-tuning our latest model enables all of these tasks, and this has interesting implications for robotics, Moravec's paradox, and the future of large models in embodied AI.
More below!
We discovered an emergent property of VLAs like Ο0/Ο0.5/Ο0.6: as we scale up pre-training, the model learns to align human videos and robot data!
This gives us a simple way to leverage human videos. Once Ο0.5 knows how to control robots, it can naturally learn from human video.
operationalizing the work behind this release was a complex real-world challenge. a huge thanks to all our operators who made it happen with thoughtful, creative execution at every step :)
Our model can now learn from its own experience with RL! Our new Ο*0.6 model can more than double throughput over a base model trained without RL, and can perform real-world tasks: making espresso drinks, folding diverse laundry, and assembling boxes.
More in the thread below.
We got a robot to clean up homes that were never seen in its training data! Our new model, Ο-0.5, aims to tackle open-world generalization.
We took our robot into homes that were not in the training data and asked it to clean kitchens and bedrooms. More below‡οΈ
Chicken, bacon & leek pie is one of my all-time favourite meals! Itβs cheap and easy to make, tastes delicious, and tastes like love.
In this thread Iβll show you how to make it!
Here's the recipe and if you don't believe me that sweet potato is really the better option, you can sub in pumpkin in the same quantity: https://t.co/t0Dt9q0Vk9
At Physical Intelligence (Ο) our mission is to bring general-purpose AI into the physical world.
We're excited to show the first step towards this mission - our first generalist model Οβ π§ π€
Paper, blog, uncut videos: https://t.co/XZ4Luk8Dci