Google DeepMind Topic

Google introduced Gemini 1.5 Pro with an experimental one-million-token context window and a Mixture-of-Experts architecture.

SELF-DISCOVER Lets Language Models Design Their Own Reasoning Structure

Google DeepMind and USC researchers presented a prompting framework that builds task-specific reasoning structures before producing an answer.

Google introduced Gemini in Ultra, Pro and Nano versions, designed natively to work across text, code, audio, images and video.