Meta Releases 5 New AI Models Which Can Accelerate Future Research

Image source: Analytics Drift

Recently, the Meta Fundamental Research Team publicly released five new AI models, including image-to-text and text-to-music generation.

Image source: Meta

Meta’s FAIR Team is Releasing New AI Models

Chameleon is a family of mixed models that can understand and generate images and text. It takes any combination of text and image as input and also outputs a combination of text and image.

Image source: AD

Meta Chameleon Models

Meta’s multi-token prediction is a better and faster large language model, which can predict multiple future words at once.

Image source: Meta

The Multi-Token Prediction: Faster Language Traning Model

Meta’s JASCO offers more control over AI music generation. It can accept various inputs, such as cords or beats, to improve the output of generated music.

Image source: AD

JASCO:  Music Generation Model

AudioSeal can be the first audio watermarking technique. Its detection speed is 485 times faster than other methods, helping prevent misuse of speech-related generative AI.

Image source: AD

AudioSeal: Detect AI-Generated Speech

Meta has developed automatic indicators to evaluate potential geographical disparities to ensure fairness and representation in text-to-image models.

Image source: Canva

Automatic Indicators: Improve Text-To-Image Diversity

The release of Meta’s new AI generative models is an important milestone and significant step in the field of AI. It shows the potential for AI to enhance productivity and drive innovation.

Image source: AD

Advancing Artificial Intelligence

Get the latest updates on AI developments


Join our

Channel Now!

Produced by: Saloni Agrawal Designed by: Prathamesh