🐰 Meta Cranking out the AI Models

Author

bunny pixel
April 29, 2024

AI continues its rapid advancement across multiple fronts. In this edition of the BunnyPixel newsletter, we'll break down the latest AI news and breakthroughs with a few fun bits along the way. Lets get started…

Meta Announces Meditron

We have been talking about the coming bloom of industry specific AI models for a while now. Medicine is one of the first obvious ones and there are many startups trying to fill this space.

Meta just came out of left field with an open source LLM AI model for low resource medical settings.

Wow. Open source and free.

We knew this is a lucrative market so it is a surprise for Meta to come out with an open source model so quickly.

Great news for the health industry and hopefully it helps improve health services for people worldwide.

AI Art

Give the post a like if you dig it !

More AI from Meta - LLama3

Meta is really working hard and cranking out some AI tech for you. They have unveiled Llama 3, a groundbreaking open-source language model with advanced capabilities previously available only in proprietary AI.This one is a ChatGPT competitor but it is free and open source. So far is right up at the top of the rankings, even close to ChatGPT4.

With models featuring up to 70 billion parameters and trained on over 15 trillion tokens, Llama 3 excels in a range of AI tasks, from logical reasoning to code generation. It's built on a decoder-only transformer architecture for high efficiency and includes safety tools like Llama Guard and Cybersec Eval to prevent misuse.

Available now on platforms like Databricks, Groq and Google Cloud, Llama 3 is setting new standards for accessible, high-performance AI technologies, promising future enhancements that include multilingual support and multimodal features.

Also, in the same week. Mistral came out with their newest AI model, also claiming it is the best open source AI Model.

AI meme of the day

Share and give the post a like if you dig it !

Microsoft’s Amazing Photo to Talking Head

Microsoft has unveiled VASA-1, an innovative AI technology capable of creating realistic videos of talking faces from just a single image and audio clip.

It offers features, such as lifelike facial expressions, head movements, and even the ability to adjust the gaze and spatial perception of the avatar.

The quality is truly amazing, the best we have ever seen. Seriously, go check out the videos in the article.

They have said they aren’t releasing it yet, until they feel it is safe to do so. Our take on that is that it will be after the American election in the fall / winter.

Bits and Bytes

AI Quote of the Day

Share and give the post a like if you dig it ! Or leave a comment if you have thoughts about it.

And that’s a wrap for today. See you in the next issue…

We would love to hear your comments, let us know what you think of the newsletter and what you would like to see in the future. Just post a comment in the community (link below)

The BunnyPixel Team
🐰

Come and check out our free AI training portal and community of business owners, entrepreneurs, executives and creators. Also, level up your skills and business with our free online AI training courses. New courses added weekly.