Gemini: A Family of Highly Capable Multimodal Models

arXiv:2312.11805v2 Announce Type: replace-cross
Abstract: This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first …

abstract applications arxiv audio capabilities cases cs.ai cs.cl cs.cv evaluation family gemini image memory multimodal multimodal models reasoning report tasks text text understanding type understanding video