← Back to feed

Google DeepMind’s Gemini: the unified multimodal answer

Google merged Brain/DeepMind efforts into Gemini, a multimodal family natively trained for text, code, and image.

Google DeepMind’s Gemini: the unified multimodal answer

In December 2023, Google presented Gemini, the result of merging Google Brain and DeepMind. The pitch: native multimodality and TPU-scale infrastructure.

The launch had demo stumbles, but Ultra/Pro/Nano covered cloud and mobile. Bard was rebranded under Gemini.

Gemini reaffirmed that the frontier race is also a distribution race (Search, Android, Cloud).