From Gemini 2.5 models to Flow: Check announcements at the Google I/O 2025

Google has made some exciting announcements at the Google I/O 2025 and this includes some interesting innovations. Most of the announcements are focused on AI. The announcements include new features for Gemini 2.5 models, Imagen 4 image generator and much more.

Some of the important announcements from the Google I/O 2025 have been mentioned below.

Flow which is Google’s new AI filmmaking tool can help storytellers to create mind-blowing cinematic clips. This tool combines the capabilities of Veo, Imagen and Lyria models for offering detailed cinematic scenes. The users can control camera motion, angles as well as perspective of their videos and they can edit and extend previously generated videos. Google AI Pro and Ultra subscribers in the US can now access Google Flow.

Google’s latest AI video model- Veo 3 gets features like improved text-to-video prompt recognition. The video model can make videos with character dialogue, background noises and sound. Google AI Pro and Ultra subscribers in the US can now access Google Flow. Veo 2 gets camera movements as well as object addition and removal functionality.

Imagen 4 which is Google’s newest AI image generator can make images up to resolution of 2K. The company has claimed that Imagen 4 offers more accuracy in terms of generating cards, posters as well as comics. It is available across Gemini, Google Workspace, Whisk and Vertex AI apps.

Gemini 2.5 offers extended language support for more than 24 languages. This comes with support for text-to-speech with expressive voices. Google has claimed that this improves things like reasoning, multimodality, coding and much more on Flash as well as Pro models.