GPT-6 Sol and Luna

OpenAI has introduced GPT-6 Sol and Luna, two large language models that have achieved state-of-the-art performance in various natural language processing tasks. The models were trained on a dataset of 1.5 trillion parameters, which is 2.5 times larger than the dataset used to train the previous GPT-6 model. Sol and Luna have achieved a score of 148 on the SuperGLUE benchmark, surpassing the previous record of 146 set by GPT-6. The models have also demonstrated strong performance in other benchmarks, including the GLUE benchmark and the Winograd Schema Challenge.