This repository contains a CTM trained on ImageNet, as described in our paper Continuous Thought Machines.

This CTM was trained on ImageNet-1K classification, demonstrating the architecture's ability to leverage neural dynamics for visual reasoning. The model exhibits unique emergent properties including adaptive computation (dynamically adjusting internal processing based on input difficulty) and interpretable attention patterns where it learns to "look around" images before making predictions, entirely without explicit training signals for this behavior. The model achieves strong calibration across its iterative refinement process.

This model is provided for research and development purposes only and should be considered as an experimental prototype. It is not intended for commercial use or deployment in mission-critical environments. Use of this model is at the user's own risk, and its performance and outcomes are not guaranteed. Sakana AI shall not be liable for any direct, indirect, special, incidental, or consequential damages, or any loss arising from the use of this model, regardless of the results obtained. Users must fully understand the risks associated with the use of this model and use it at their own discretion.