Want to save the planet from AI? Chuck in an FPGA and ditch the matrix

Watts down, doc: Boffins find machine learning models can function with more modest power requirements

Large language models can be made 50 times more energy efficient with alternative math and custom hardware, claim researchers at University of California Santa Cruz.…

alternative california claim custom energy energy efficient fpga function hardware language language models large language large language models machine machine learning machine learning models math matrix planet power requirements researchers santa save the matrix university university of california