Our benchmark indicates that for training 200 Million parameter neural network, ThirdAI’s BOLT Engine matches (even surpasses) the acceleration of an A100 GPU while running on Apple M1, an 8 core notebook. It should be noted that a new M1 laptop costs only 1/10th and consumes 1/4th the power of an A100. Please refer to comparison for detailed benchmarks.
Third-AI builds hash-based processing algorithms that accelerate the training and inference of neural networks. The technology is a result of 10 years of innovation in finding fundamental ways to make deep learning more efficient.
USEFUL LINKS
Technology
Contact