There are ways to manage the non-deterministic nature of LLMs. Each of these approaches has its own trade-offs.

Critics of LLM-based products like ChatGPT, Claude, Midjourney, and other such products like to brush them off as just this year’s version of NFTs. They’re crypto bullshit being peddled by the same jokers who are just out there to stow disinformation and make a quick buck. I won’

Some explanations should be written by code authors. Others may best be generated on the fly by LLM-assisted code readers.

There are many benefits of running a private LLM for your company or product, but it all boils down to being able to provide real-time data in context.

ChatGPT and LLMs can do anything, so what can you do with them? How do you know? Do we move to chat bots as a magical general-purpose interface, or do we unbundle them back into single-purpose software?

The open source LangStream framework combines data streaming with generative AI. We speak with project lead Chris Bartholomew from DataStax.

Jon Udell uses ChatGPT, Cody and GitHub Copilot to help him build an ODBC Plugin for Steampipe, an extensible SQL interface to cloud APIs.

AI presents challenges for data privacy in Large Language Models (LLMs) like ChatGPT emphasizing the need for robust security measures.

LLM Guard is a toolkit designed to fortify the security of Large Language Models, and it's freely available for usage with various LLMs.

As websites start to block Common Crawl, and as the project leans in to its role in training LLMs, will it become harder to use data from the web for other purposes?

Prompt Engineering, also known as In-Context Prompting, refers to methods for how to communicate with LLM to steer its behavior for desired outcomes without updating the model weights. It is an empirical science and the effect of prompt engineering methods can vary a lot among models, thus requiring heavy experimentation and heuristics. This post only focuses on prompt engineering for autoregressive language models, so nothing with Cloze tests, image generation or multimodality models.

A paper has shown that a compression algorithm – gzip – outperforms some large language models (LLMs) in some tasks. This has the NLP community …

LLMs can deliver just-in-time knowledge tailored to real programming tasks, making it a great way to learn about coding idioms and libraries.

A primer on what large language models are, why they are used, the different types, and what the future may hold for LLM applications.

In the LLM app stack, a playground is where developers can test out (and deploy) prompts. We discussed this new concept with Humanloop's CEO.

Web3 failed to remake the developer ecosystem, but the emerging LLM stack is forcing the cloud native era to adapt. We examine its progress.

To train a large language model, you must overcome three big challenges: data, hardware and legal. It helps to be a large organization, too.

LLMs hold the key to generative AI, but some are more suited than others to specific tasks. Here's a guide to the five most powerful and how to use them.

A successful first use of GPT-4 Code Interpreter raises the hope that LLMs can help democratize scripting.

🤖 What are LLMs? How can you use them? And why should you care?

In the 20th episode of my Mastodon series I pivoted to a new topic: LLM-assisted coding. After three posts in the new series, it got picked up by The New Stack. Here’s the full list so far, I…