Google announced the launch of its latest and most powerful AI model called Gemini, which comes in three sizes - Gemini Ultra, Gemini Pro, and Gemini Nano. Gemini Ultra is said to be the first model to outperform human experts on a massive multitask language understanding test involving 57 subjects. Gemini will be licensed through Google Cloud and also power Google products like its Bard chatbot and Search Generative Experience. Google says Gemini Pro outperformed OpenAI's GPT-3.5 on tests but did not provide comparisons to GPT-4. The company also introduced its next-gen TPU v5p chip for training AI models, though performance comparisons to Nvidia were not given.