Paul Yang

Discover key AI components for building intelligent applications, from language models for planning to classifiers, retrieval models, and response generators. Learn how models like GPT-o1, BERT, and CLIP enhance AI agents.

Components of a Compound AI System

Use Runhouse with marimo to enable a great development experience and reproducible execution without being limited by the scale of compute.

Marimo x Runhouse for Reproducible ML

How Does Cross Entropy Loss Work in PyTorch?

Flux1 is a new generative AI model that lets you create high-quality images based on text input prompts. We will show you how to easily host the model on your infrastructure, using AWS in this example. 

How to Run and Host Flux.1 Image Generation (on your own cloud)

Discover the complete guide about how to host your own image generation AI with StableDiffusion. This comprehensive blog post walks you through the steps of setting up StableDiffusion XL on AWS. Learn about the hardware requirements, software dependencies, and configuration process needed to launch your own image AI.

StableDiffusion XL: How to Host Your Own Image Generation AI

Llama 3.1 is the newest in the family of Facebook Large Language Models (LLMs). Its performance is comparable to best in class closed source models, and this guide will show you how to run it on your own infrastructure. 

How to Deploy Llama 3.1 To Your Own Infrastructure (AWS Example, Released July 2024)

AWS SageMaker is a comprehensive platform for ML. Who should use it and why? And why not use it? We cover in detail the benefits and the components of the platform, but also the cons. 

What is Amazon SageMaker? What Are the Alternatives to Sagemaker?

Matt Kandler

Learn to host your own LLM with vLLM, deploying Llama 3 on any cloud quickly and cost-effectively. vLLM enables faster and more efficient inference, while Runhouse lets you deploy your code instantly to any cloud compute with no overhead. 

How to Host Your Own LLM with vLLM: The Cheapest, Fastest, and Easiest Way 

LoRA, or Low-Rank Adaptation, is a technique used to fine-tune large language models more efficiently. To understand LoRA at a CS 101 level, let's break down the concept first, before sharing a coding example that lets you get started with fine-tuning LoRA on any compute.

Low Rank Adaptaion or LoRA is a fine-tuning method. A head with gears in it to represent fine tuning

Guides