I wish I was kidding. I made this image with Google's Nano Banana. Then I accidentally created a world around it. 🍌 Left: Slick, a retired war-bot with a futuristic gun and a Banana. 🍎 Right: Tommy Appleseed, an apple-shaped assassin with a sword and trust issues. What

Google is improving their AI suitcase. They are taking the lead in the AI race step by step. - 100 free credits for every new user per month - this equals 1 quality or 5 fast video generations in Veo 3 Time for some experiments 🤔🧐

🚨BREAKING:🚨 This is an insane move by Google! All Google AI Ultra Subscribers are getting free unlimited fast video generations in Veo 3! (forever?) Creators have an endless opportunity to let their imagination FLOW now! Thank you so much! FlowbyGoogle Google DeepMind

Google is at it again! They just released Google Vids, a platform for video creators inside Google Workspace. Only available in the U.S. for now! 🇺🇸📹🧑‍🎨

You don’t need to break the bank creating a promo video for your brand Yesterday I showed you how to create the stills, now im showing you how to make the video This 10s fashion ad was built from stills to full motion video Stills: Gemini 2.5 Flash (aka nano-banana) - images

This Isometric figures trend is blowing up right now 🔥

Wow, look at these amazing results using this prompt: Generate a photo of a girl cosplaying this illustration, with the background set at Comiket. Try it out in Nano Banana 🍌

This week was…bananas 🍌. Here’s what launched: — We released Gemini 2.5 Flash Image, our latest state-of-the-art image generation and editing model in the Google Gemini App, @GoogleAIStudio for developers, and @GoogleCloud Vertex AI for enterprise. —@JoshWoodward, Jeff Dean, and

Very helpful tutorial for video creation in Veo 3.

🇨🇳 Another great Chinese Model, OmniHuman-1.5 from ByteDance Turns 1 image plus a voice track into expressive avatar video by pairing a System 1 and System 2 inspired planner with a Diffusion Transformer, Produces coherent motion for over 1 minute with moving camera and multi

🚨 Breaking: Prompt → playable 3D world is live. Tencent’s open-source HunyuanWorld turns one line of text into an explorable 3D scene with exportable meshes, separated objects, 4090-ready. The worldbuilder is now a prompt. 🚀

Guys, don't sleep on open-source models! Congratulations on the launch Kimi.ai I'm really excited to see the front-end coding capabilities of this model.

Creating TikTok content just became effortless. Mirage - first AI built purely for short-form content: - Talking humans generated on demand - Native 9:16 format understanding - Short-form pacing built-in We just unlocked infinite content creation. The bottleneck was never