- Brain Scriblr
- Posts
- LLAMA 3.1 from Meta is here
In partnership with
News
Salesforce has announced a new AI Service Agent for its platform, using their Einstein Copilot as the foundation. It can autonomously handle customer service tasks and integrate with company records for seamless support.
Andrej Karpathy, legendary researcher, has announced the creation of Eureka Labs, an AI-native school. Andrej is known for being extremely educational about AI, and this company could personify that mission.
As for our weekly touchpoint on OpenAI, anonymous ex-OpenAI whistleblowers have accused OpenAI of using NDAs that illegally restrict employees from reporting issues to regulators.
I will also mention that there are rumors surrounding OpenAI and a potential new product titled Strawberry. At this point there is a lot of speculation but not many facts. I try to stick to the facts.
Your Brilliant Business Idea Just Got a New Best Friend
Got a business idea? Any idea? We're not picky. Big, small, "I thought of this in the shower" type stuff–we want it all. Whether you're dreaming of building an empire or just figuring out how to stop shuffling spreadsheets, we're here for it.
Our AI Ideas Generator asks you 3 questions and emails you a custom-built report of AI-powered solutions unique to your business.
Imagine having a hyper-intelligent, never-sleeps, doesn't-need-coffee AI solutions machine at your beck and call. That's our AI Ideas Generator. It takes your business conundrum, shakes it up with some LLM magic and–voila!--emails you a bespoke report of AI-powered solutions.
Outsmart, Outpace, Outdo: Whether you're aiming to leapfrog the competition or just be best-in-class in your industry, our custom AI solutions have you covered.
Ready to turn your business into the talk of the town (or at least the water cooler)? Let's get cracking! (And yes, it’s free!)
Research
The paper KAN or MLP: A Fairer Comparison provides a detailed comparison between Kernel Attention Networks (KAN) and Multi-Layer Perceptrons (MLP) across various tasks, including machine learning, computer vision, audio processing, natural language processing, and symbolic formula representation. The study aims to offer a balanced evaluation by controlling the number of parameters and Floating Point Operations (FLOPs) for both models.
The paper INF-LLaVA, introduces a novel Multimodal Large Language Model (MLLM) designed to address the challenges of high-resolution image perception. Traditional MLLMs face limitations due to the quadratic complexity of their vision encoders, which restricts the resolution of input images. INF-LLaVA overcomes these limitations through two innovative components: the Dual-perspective Cropping Module (DCM) and the Dual-perspective Enhancement Module (DEM).
Tools
Open source tool Exo allows you to run an AI cluster at home on everyday devices.
Ubiops allows you to deploy your AI workload.
Loops helps you to identify key areas to improve company/brand KPIs.
Prompt
Prompt » younng woman reading about sustainability in front of plant stand
Non-Image Prompt
Act like an AI and business expert, and come up with a simple but unique AI business model for making money (let your ideas be so unique they can’t be found elsewhere).
Main Newsletter Topic
Meta's AI Model That's More Than Just a Wooly Idea
After months of anticipation and a dramatic leak just yesterday, Meta has officially unveiled the latest gem in its open-source LLM lineup: LLAMA-3.1 |
What's the hype about? |
It’s just a tiny tweak to the Llama 3 model, but it packs a punch with Llama 3.1 405B—a 405 billion parameter monster, the world’s largest open-source LLM to date, surpassing NVIDIA's Nemotron-4-340B-Instruct Meta claims it has outperformed GPT-4, GPT-4o, and Claude 3.5 Sonnet across various tasks. This model was trained on a whopping 15 trillion tokens using over 16,000 H100 GPUs, making it Meta’s most massive and ambitious model yet. |
How to try it? |
You can download the models directly from Meta or Huggingface. Unless you have a powerful computer with multiple cores it would not be practical to use this yourself. You would need to rely on hosted services or have a computer that runs on multiple cores. |
