Ai News: Animate-X, Adobe Firefly Latest Information And More

Animate-X is a new tool that helps turn different character images into animations. This includes human characters and animal-like characters that look human, which are often seen in games and shows.

https://lucaria-academy.github.io/Animate-X

https://github.com/Lucaria-Academy/Animate-X

Adobe has made some cool updates to Firefly, its AI tool. The new Firefly Video Model (beta) can now make videos from text and images. You can use it in Premiere Pro, which makes video editing much easier. Firefly Image 3 is faster now, making images up to four times quicker, so you can create your ideas faster. Photoshop has a new area called Generative Workspace (beta) where you can use AI to help with your designs. Illustrator’s Firefly Vector Model (beta) is also better, making it easier to create vector graphics. They added new features for businesses, like translating videos and editing lots of images at once.

Introducing Substance 3D Viewer (beta) a new 3D desktop app that can open, view, and render 70+ file formats.

In @Photoshop (beta), drag & drop a 3D model on your canvas, and it becomes an editable Smart Object.

Download it for free and give it a try! https://t.co/3qo17o87XM pic.twitter.com/HzP6E0CavV

— Adobe Substance 3D (@Substance3D) October 14, 2024

https://news.adobe.com/news/2024/10/101424-adobe-launches-firefly-video-model

www.adobe.com/products/premiere/ai-video-editing.html

https://jp.reuters.com/economy/industry/LGNGVL7VNRJZVIWSPWETPPLUQY-2024-10-15

OmniBooth is a new tool that helps you control how images are made. You can use masks (which are like stencils) and text or pictures to create images that match what you want.

https://len-li.github.io/omnibooth-web

https://github.com/EnVision-Research/OmniBooth

https://github.com/EnVision-Research/OmniBooth

CtrLoRA is a new system for making images you can control. It uses a lot of pictures to train a basic network and smaller networks for special tasks.

https://arxiv.org/abs/2410.09400

https://huggingface.co/xyfJASON/ctrlora/tree/main

https://github.com/xyfJASON/ctrlora

CogView3 and CogView-3Plus are AI models that create images from text. CogView3 makes images with better quality and uses less power, while CogView-3Plus is even better and creates images with more detail.

https://github.com/THUDM/CogView3

Cagliostro Forge Colab is a tool that helps you create images using Stable Diffusion. It’s easy to set up and start using on Google Colab.

https://github.com/cagliostrolab/forge-colab

Zamba2-7B is a new language model made by Zyphra. Even though it’s small, it works very well and is better than other models like Mistral and Llama3.

https://www.zyphra.com/post/zamba2-7b

txtai is a search database that can find not just text but also images and sounds. The new version 7.5 has been released. It can find similar items based on meaning, not just keywords.

https://github.com/neuml/txtai

ChatGPT now has a new feature called “SearchGPT,” which is available for some users.

https://chatgpt.com/share/670c2d01-1b18-8001-866d-ba9eb2803260

Google has signed a contract to build seven small nuclear reactors to power its AI data centers. These should be ready between 2030 and 2035.

https://www.engadget.com/big-tech/google-strikes-a-deal-with-a-nuclear-startup-to-power-its-ai-data-centers-201403750.html?guccounter=1

There’s a new article that compares 50 generative AIs that support the Japanese language. It says that “Gemini1.5Pro” is the best at not lying.

https://www.nikkei.com/article/DGXZQODL1858X0Y4A910C2000000

That’s all for today! I went through the news quickly because I want to try Adobe’s Firefly. I’ll check out the beta version of Photoshop and Premiere. I wonder if they will add Firefly to other apps too.

I sometimes think about how many people actually use all the Adobe CC apps. I use many of them, but I know people who only use Photoshop and Illustrator. I like to try everything, but it’s filling up my storage! See you next time!