GPT-4 Built this new multimodal model!!

GPT-4 Built this new multimodal model!!

Good morning fellow AI enthusiast! This week's iteration focuses on a very hot GitHub repository and research called LLaVA, an end-to-end large multimodal model that connects a vision encoder and LLM for general-purpose visual and language understanding.


Receive the weekly digest right in your emails ??

GPT-4 is powerful, but did you know that some AIs are built entirely thanks to it? Yes, GPT-4 is so good that it can be used to generate good enough data to train other AI models. And not any model but better models than itself!

Liu et al. just used GPT-4 to create a general-purpose language vision model called LLaVA, the first general-purpose model that understands and follows visual and language-based instructions. Yes, they didn't use GPT-4 as the base model, but to train their model! As we will see in the video, GPT-4 was used to generate a large and high-quality dataset to train a new model that understands images. Oh and obviously it not only understands images but also text (there's the multimodality), which means it can answer a wide variety of questions about them! Learn more in the full article or in the video...

We are incredibly grateful that?the newsletter?is now read by over 12'000+ incredible human beings counting our email list and LinkedIn subscribers. Reach out [email protected]?with any questions or details on sponsorships or visit my Passionfroot profile. Follow our newsletter at Towards AI, sharing the most exciting news, learning resources, articles, and memes from our Discord community weekly.

If you need more content to go through your week, check out the podcast!

Thank you for reading, and we wish you a fantastic week! Be sure to have?enough rest and sleep!

Louis


Elliott A.

Senior System Reliability Engineer / Platform Engineer

1 年

Yup, this is the way. Big time,

  • 该图片无替代文字
Janvi V.

Data Scientist/Data Analyst/Machine Learning Engineer/ Data Storyteller/SQL/Tech./"Unleashing the Power of Data: Data Scientist and AI Alchemist, Aiming to Revolutionize the Future of Analytics"

1 年

Sounds promising.

回复
Tyler Suard

Senior AI Researcher & Developer at Parker-Hannifin. Ex-Apple, Ex-Meta. Contributor to Autogen, Tensorflow, PyTorch, Huggingface Transformers. Stanford affiliate. Interested in longevity, AI +Bio.

1 年

Excellent explanation, thank you

回复

要查看或添加评论,请登录

Louis-Fran?ois Bouchard的更多文章

  • Want to start programming in the AI era? This is for you...

    Want to start programming in the AI era? This is for you...

    Good morning! If you’ve been wanting to break into AI development but feel like your coding foundation isn’t quite…

  • Using AI for Writing

    Using AI for Writing

    Good morning! We’ve (Towards AI) been using AI to research, plan, help us with drafts, and refine our lessons for our…

    4 条评论
  • How LLMs Are Changing Every Job

    How LLMs Are Changing Every Job

    Good morning! Today, I’m sharing our third video out of 6 we made for our “8-hour Generative AI Primer” course. In this…

  • LLM Developers: The future of software development

    LLM Developers: The future of software development

    Software engineers vs. ML engineers vs.

    1 条评论
  • Real Agents vs. Workflows

    Real Agents vs. Workflows

    What most people call agents aren’t agents. I’ve never really liked the term “agent”, until I saw this recent article…

    1 条评论
  • CAG vs RAG: Which One to Use?

    CAG vs RAG: Which One to Use?

    If you're using ChatGPT or other AI models, you've probably noticed they sometimes give incorrect information or…

    2 条评论
  • Why LLMs Are the Future of Work

    Why LLMs Are the Future of Work

    Good morning! Today, we start the new series of videos for our most recent Towards AI course: 8-hour Generative AI…

    1 条评论
  • Introducing Our 8-Hour Generative AI Primer

    Introducing Our 8-Hour Generative AI Primer

    Once again, I’m super excited to share some news from the Towards AI team—we’ve just launched a brand-new 8-hour…

    11 条评论
  • Best Practices for Building and Deploying Scalable APIs in 2025

    Best Practices for Building and Deploying Scalable APIs in 2025

    Good morning! When we talk about building powerful machine learning solutions, like large language models or…

    1 条评论
  • Lessons from Nvidia Minitron

    Lessons from Nvidia Minitron

    We’d all love to build a model from scratch like Llama, but how realistic is that? The computing, architecture, and…

    3 条评论

社区洞察