Welcome to the Predibase AI Resource Center

Your one stop shop for all LLM, generative AI and machine learning tutorials, ebooks, webinars and learning resources.

DeployCon: A Summit for Teams Serious about Production AI
DeployCon is an engineer-first summit for teams building the next generation of AI. The top minds in LLMOps from Pinterest, DoorDash, Tinder, Meta share their strategies for deploying, scaling, and continuously improving AI systems in production.
Videos & Webinars
Precision Vision: Supercharging Vision Language Models with Fine-Tuning
Watch on-demand webinar: Supercharging Vision Language Models with Fine-Tuning
Videos & Webinars
The Fastest Way to Serve Open-Source Models: Inference Engine 2.0
Watch on-demand webinar: How to serve open-source models 2–4x faster—no tuning, no headaches.
Videos & Webinars
Shared vs. Private LLMs: Comparing Cost, Latency and Capabilities
Deep dive session why shared LLMs fail at scale, and how to adopt fast efficient private LLMs
Videos & Webinars
Build vs Buy: Choosing the Right AI Stack
Explore the hidden costs of DIY AI, pro/cons of building vs. buying AI infra and how to assess your readiness.
eBooks
Complete Guide to Reinforcement Fine-Tuning
Read the complete Guide to RFT to learn how to achieve 20%+ higher accuracy than GPT-4 using minimal labeled data including use cases and tutorials.
eBooks
Intro to Predibase's Reinforcement Fine-Tuning Platform
Deep dive webinar on the first end-to-end platform for training and serving LLMs with RFT. Benchmarks and example code included.
Videos & Webinars
DeepSeek Adoption Survey and Infographic
We survey 500+ AI professionals to separate the reality from the hype, explore use cases and adoption of DeepSeek.
eBooks
Complete Guide to Serving LLMs
Deep dive into 3 essential strategies for mastering fast, efficient and cost-effective LLM deployments.
eBooks
Case Study: Checkr
Learn how Checkr reduced OpenAI costs by 5x and improved inference 30x with fine-tuned SLMs.
eBooks
Reinforcement Fine-tuning for DeepSeek
In this tech talk and demo, you'll learn how to fine-tune DeepSeek-R1 with reinforcement learning.
Videos & Webinars
What is GRPO and Reward Functions
AMA style demo on GRPO fine-tuning and a deep dive on creating reward functions.
Videos & Webinars
Demo: Predibase Reinforcement Fine-tuning Platform
Introducing the first end-to-end platform for Reinforcement Fine-Tuning (RFT).
Videos & Webinars
Turbo Reasoning: How to Accelerate Models like DeepSeek
Learn how to increase the throughput of reasoning models by 2-4x without sacrificing accuracy.
Videos & Webinars
Customer Talk: MMC's Agentic AI Assistant
How Marsh McLennan built an AI assistant with fine-tuned SLMs that saves them millions of hours.
Videos & Webinars
SmallCon: Virtual Conference for GenAI Builders
Talks from HuggingFace, Nvidia, Salesforce and more on practical strategies for small models.
Videos & Webinars
Optimize Inference for Fine-tuned SLMs
Learn how to speed up accelerate your LLM inference by 4x, improve reliability, and reduce costs with FP8, Turbo LoRA and more.
Videos & Webinars
Solution Guide: Developer Platform for Productionizing Open-source AI
Learn about the fastest and most efficient way to fine-tune and serve pre-trained open-source AI models.
eBooks
Customer Talk: Automating Customer Service
How Convirza analyzes millions of calls with SLMs and multi-lora serving.
Videos & Webinars
Notebook: Fine-tune a SLM for Content Summarization
Learn how to easily fine-tune Llama-3.1-8B to accurately summarize a chat conversations.
Tutorials
Predibase Wrapped: Hits from 2024
Showcase of our biggest AI innovations: continued training, fine-tuning VLMs, Turbo LoRA and more.
Videos & Webinars
Customer Talk: Clearwater Builds AI Agents with SLMs
Learn how Clearwater Analytics built and deployed an agentic workflow for their customers.
Videos & Webinars
Beat GPT-4 with a Small Model and 10 Rows of Data
Learn how to leverage large language models like Llama-3.1-405b to generate synthetic data to fine-tune a small model that achieves GPT-4 level results.
Videos & Webinars
Your Models, Your Cloud: Secure LLMs in Your VPC
Learn about different approaches to securing GenAI workloads and how to customize and host LLMs in your private cloud the easy way.
Videos & Webinars
Customer Talk: Checkr
Learn how Checkr uses Predibase to fine-tune and serve LLM classifiers to automate the background check process—with higher accuracy and a 5x lower cost than OpenAI.
Videos & Webinars
Notebook: Build a SQL CoPilot
This example notebook will show you how to fine-tuned Llama-3.1-8b on a high-quality synthetic SQL dataset to create a highly accuracy code generation model.
Tutorials
Demo: Synthetic Data Generation
See how you can generate high quality synthetic data that can then be used to instantly fine-tune your model all within Predibase.
Videos & Webinars
Notebook: Customer Support
In this notebook tutorial, we'll show you how to fine-tune Mistral-7B for 3 different customer support use cases and serve on a single GPU with Predibase and LoRAX.
Tutorials
The Definitive Guide to Fine-tuning LLMs
This guidebook provides practical advice for overcoming the four primary challenges teams face when fine-tuning from managing infra to understanding techniques for efficient fine-tuning.
eBooks
Notebook: Docstring Generation Using CodeLlama-13b
This tutorial notebook will help you get started fine-tuning CodeLlama-13b on predibase to generate code documentation.
Tutorials
Graduate from OpenAI to Open-Source
Watch this on-demand session to learn our 12 best practices for distilling smaller LLMs with GPT.
Videos & Webinars
How–and when–to fine-tune and serve LLMs
Learn how to quickly and effectively build end-to-end workflows to customize open-source LLMs on your data in your own cloud incl. best practices like LoRA and quantization.
Videos & Webinars
Serve 100s of Fine-tuned LLMs for the Cost of Serving One
Learn about LoRAX the new open-source framework that makes it possible to dynamically serve many LLMs on a single GPU without any degradation to performance.
Videos & Webinars
Survey Report: A Look at LLMs in Production
Insights from 150 developers and data scientists on what it takes to be successful with LLMs, the challenges that still exist, and recommendations moving forward.
eBooks
Notebook: Fine-tune a Code Generation LLM with Ludwig
This detailed tutorial notebook will show you how to use the new features in open-source Ludwig 0.8 to deploy and fine-tune the Llama-2 LLM for a code generation task.
Tutorials
Panel Discussion: Building Custom LLMs in Production
Interactive discussion with LLM experts from Uber, LLamaIndex, Bank of America, and Predibase on avoiding common pitfalls when building production-grade LLM apps, tips for customizing and serving models, and future of GenAI.
Videos & Webinars
ML Real Talk: 10 Things You Must Know About LLMs
In this episode of ML Real Talk, Host Daliana Liu and, Arnav Garg, ML Engineer, delve into all things LLMs from how these complex models are trained to exploring LLM customization like few shot learning to finetuning.
Videos & Webinars
Efficiently Build Custom LLMs on Your Data with Open-source Ludwig
With Ludwig v0.8, we’ve released the first open-source, low-code framework optimized for efficiently building custom LLMs on private data. Learn how to use it.
Videos & Webinars
How-to Guide: Overcoming Overfit Models
Learn best practices for how to address and avoid overfitting in your ML models.
Tutorials
Build Your Own LLM in Less Than 10 Lines of YAML
Generalized models solve general problems. The real value comes from training a large language model on you data and finetuning it for a specific ML task. Now you can build your own custom LLM in a few lines of code.
Videos & Webinars
4 Ways Developers Can Build AI-Powered Apps Faster with Declarative ML
New declarative machine learning tools make ML accessible to developers by simplifying model development and deployment with a config-driven approach rooted in engineering best practices. Read the ebook to learn more.
eBooks
Privately Host and Customize Large Language Models for Your ML Task
There’s a better way to use and finetune LLMs on your own data that is faster, cheaper, and doesn’t require giving away any proprietary data to LLM vendors. Check out this webinar and demo to learn more.
Videos & Webinars
AutoML 2.0: The future of AutoML
A new generation of declarative ML systems—like those pioneered at Uber, Apple, and Meta— provide a glass-box approach to automating ML that enables teams to bring new models to market faster with complete flexibility and control.
Videos & Webinars
Hands-on with Multi-Modal Machine Learning: Predicting Customer Ratings
Learn how to easily build and evaluate a deep learning model to predict customer ratings using unstructured text and tabular data.
Videos & Webinars
Santavision: Deep Learning for the Holidays
Learn how to use Predibase and open-source Ludwig to rapidly train an image classifier in less than 15 lines of a config file
Videos & Webinars
Intro to Text Classification on Ludwig
Learn how to build a state-of-the-art text classifier with Ludwig in just a few lines of a configuration file.
Tutorials
Exploring AutoML for Deep Learning with Ludwig
Ludwig AutoML makes it easy to rapidly build state-of-the-art models for tabular and text classification among other use cases. In this session, we deep-dive into Ludwig AutoML design, development, evaluation, and use.
Videos & Webinars
Intro to Declarative ML Systems
Learn how Declarative machine learning systems—like those started at Uber and Apple—provide a glass-box approach to automating ML.
Videos & Webinars
Guide to Sentiment Analysis with Ludwig
3-part guide to go from importing data and implementing baseline models to working with BERT and doing Hyperparameter Optimization, all using only open-source Ludwig.
Tutorials

Ready to efficiently fine-tune and serve your own LLM?

Try Predibase for Free