Gemini (OpenAI)
1 minute read ∼ Filed in : A paper noteIntroduction
it is Multimodal model, and is trained jointly across image, audio, video, and text data for the purpose of building a model with both strong generalist capabilities across
Gemini models build on top of Transformer decoders, that are enhanced with improvements in architecture and model optimizations to enable stable training at scale and optimized inference on Google Tensor Proessing unit.