Excited to release FAST, our new robot action tokenizer! 🤖 Some highlights: - Simple autoregressive VLAs match diffusion VLA performance - Trains up to 5x faster - Works on all robot datasets we tested - First VLAs that work out-of-the-box in new environments! 🧵/

The key idea: FAST compresses actions before training on them. This removes redundancy & makes autoregressive VLA training on high-frequency tasks possible, where models like OpenVLA failed. We use the discrete cosine transform for compressing actions (also used by eg JPEG). 2/

With FAST, we scale autoregressive VLA training to pi0 scale, and we can solve some pretty complex robot tasks, simply via next token prediction! The best part: in our experiments, pi0+FAST converges 5x faster than diffusion pi0! Days instead of weeks of training! 🎉 3/

My favorite result: with FAST we can finally train VLAs on the DROID dataset & they work zero-shot in many scenes! Below is the same policy controlling robots at Berkeley, Stanford and UW. Just point a camera at the scene, type out an instruction, et voila! 4/

We are releasing a FAST tokenizer we pre-trained on 1M real robot action sequences. In our tests it works well across all kind of robots — and it’s all on HuggingFace! Happy VLA training! :) 5/

I am very excited about FAST, because (1) it makes VLA training really easy, even on complex tasks, and (2) with FAST it’s trivial to interleave non-robot data in VLA training (web data, subgoals, video prediction etc), it’s all just tokens! Lots of things to explore! :) 6/

Please find more details about FAST in our paper! Thanks to @KyleStachowicz and many colleagues @physical_int who helped with this project! Paper: Website:

Excited about the latest tech for your drone product? Our NDAA-compliant, US-made flight controllers are designed to accelerate your path to market and provide a solid platform for developing your autonomous software. Check them out! #Drones #UAV #UAS #Robotics #MadeInUSA

Neat idea, great work @KarlPertsch and co!

Thanks Ted! :)

Really cool paper! Congrats Karl!

Thanks Cheng! :)