Artificial Intelligence #245

Hey, in this issue: Llama 3.2: new small LLMs and vision LLMs; why aren’t humanoids in our homes yet?; training a 70B LLM from scratch from bare metal; how to finetune multimodal models or VLMs with Hugging Face; a tool for extracting structured data from PDF; AI research uncovers 300 ancient etchings in Peru's Nazca desert; and more.

The sponsors of this issue are AtScale and Galileo.

• [Meta] Llama 3.2: New small LLMs and vision LLMs
• [Application] AI research uncovers 300 ancient etchings in Peru's Nazca desert
• [Nature] A foundation model for clinician-centered drug repurposing
• [Sponsored] Unlock Business Context with GenAI: Insights from GigaOm Report

Get ahead of data trends with insights from the GigaOm 2024 Semantic Layer Report! Learn how AtScale’s semantic models empower GenAI and support decentralized teams with innovative features like the "one-click model." Join Andrew Brust and Dave Mariani for this exclusive webinar!

• [Robots] ‘It’s the robot we were all expecting – like C3PO’: why aren’t humanoids in our homes yet?
• [Explained] From bare metal to a 70B model: infrastructure set-up and scripts
• [Tool] torchao: PyTorch native architecture optimization
• [Sponsored] Introducing a new free eBook by Galileo – Mastering RAG: a developer's guide to enterprise-grade RAG systems!
• [Tutorial] How to finetune multimodal models or VLMs with Hugging Face TRL
• [Tool] MinerU: extract structured data from PDF

More than 850,000 subscribers are reading this newsletter. If you are building an AI or a data product or service, you can become a sponsor of one of the future newsletter issues and get your business featured in the newsletter. Feel free to reach out to [email protected] for more details on sponsorships.

Enjoy the newsletter? Please help us make it bigger and better by sharing it with colleagues and friends.