# Multimodal Robot Control

Static quantized version of STEVE-R1-7B-SFT, supporting multiple quantization levels for different hardware requirements

MiniVLA is a more compact yet higher-performing vision-language-action model, compatible with the Prismatic VLMs project codebase.

RDT-170M is a 170-million-parameter imitation learning diffusion Transformer model designed for robot vision-language-action tasks.

A 1-billion-parameter imitation learning diffusion Transformer model pretrained on 1M+ multi-robot operation data, supporting multi-view visual-language-action prediction

Octo Small is a robot control model trained based on diffusion policy, capable of predicting 7-dimensional actions for the next 4 steps, suitable for multi-source robot datasets.

Octo is a robot control foundation model trained on diffusion policy, capable of predicting future actions and processing multimodal inputs.