Turbocharged Training: Optimizing the Databricks Mosaic AI Stack With FP8
At Databricks, we believe that the best companies in the world, in every sector, will have AI-powered systems that are trained and customized...
Fast, Secure and Reliable: Enterprise-grade LLM Inference
Introduction After a whirlwind year of developments in 2023, many enterprises are eager to adopt increasingly capable generative AI models to supercharge their...
Fine-Grained Human Feedback
(This post written in collaboration with Zeqiu (Ellen) Wu and Yushi Hu , both PhD students affiliated with the University of Washington, and...
LIMIT: Less Is More for Instruction Tuning
How should you finetune a large language model for general-purpose question answering? One intriguing approach is that of supervised finetuning on a small...
US Air Force Hackathon: How Large Language Models Will Revolutionize USAF Flight Test
[DISTRIBUTION STATEMENT A. Approved for public release; Distribution is unlimited 412TW-PA-24004] The views expressed are those of the author and do not reflect...
OLMo Is Here, Powered by Databricks
As Chief Scientist (Neural Networks) at Databricks, I lead our research team toward the goal of giving everyone the ability to build and...
Serving Quantized LLMs on NVIDIA H100 Tensor Core GPUs
Quantization is a technique for making machine learning models smaller and faster. We quantize Llama2-70B-Chat, producing an equivalent-quality model that generates 2.2x more...
Building and Customizing GenAI with Databricks: LLMs and Beyond
Generative AI has opened new worlds of possibilities for businesses and is being emphatically embraced across organizations. According to a recent MIT Tech...
LLM Training and Inference with Intel Gaudi 2 AI Accelerators
At Databricks, we want to help our customers build and deploy generative AI applications on their own data without sacrificing data privacy or...
Integrating NVIDIA TensorRT-LLM with the Databricks Inference Stack
Over the past six months, we've been working with NVIDIA to get the most out of their new TensorRT-LLM library. TensorRT-LLM provides an easy-to-use Python interface to integrate with a web server for fast, efficient inference performance with LLMs. In this post, we're highlighting some key areas where our collaboration with NVIDIA has been particularly important.