Large Language Models

Khan Academy explores the potential for GPT-4 in a limited pilot program.

Implementing HNSW + Product Quantization (PQ) vector compression in Weaviate.

Our mission is to ensure that artificial general intelligence—AI systems that are generally smarter than humans—benefits all of humanity.

LangChain is one of the most exciting new tools in AI. It helps overcome many limitations of LLMs, such as hallucination and limited input lengths.

We’re clarifying how ChatGPT’s behavior is shaped and our plans for improving that behavior, allowing more user customization, and getting more public input int

It's like DreamBooth, but much faster. And you can run it in the cloud on Replicate.

We’re launching a pilot subscription plan for ChatGPT, a conversational AI that can chat with you, answer follow-up questions, and challenge incorrect assumptio

We’re launching a classifier trained to distinguish between AI-written and human-written text.

Many new Transformer architecture improvements have been proposed since my last post on “The Transformer Family” about three years ago. Here I did a

We’re happy to announce that OpenAI and Microsoft are extending our partnership.

OpenAI researchers collaborated with Georgetown University’s Center for Security and Emerging Technology and the Stanford Internet Observatory to investigate ho

[Updated on 2023-01-24: add a small section on Distillation .] Large transformer models are mainstream nowadays, creating SoTA results for a variety of tasks. T

Using GPT-3 to deliver fast, nuanced insights from customer feedback.

Fine-tuning GPT-3 to power and scale done-for-you video creation.

Using GPT-3 to create the next generation of AI-powered characters.

Lilian Weng works on Applied AI Research at OpenAI.

We are excited to announce a new embedding model which is significantly more capable, cost effective, and simpler to use.

The post A conversation with Kevin Scott: What’s next in AI appeared first on The AI Blog .

Learn how to import and query the Sphere dataset in Weaviate!

We’ve trained a model called ChatGPT which interacts in a conversational way. The dialogue format makes it possible for ChatGPT to answer followup questions, ad

With just a handful of images and a single API call, you can train a model, publish it to Replicate, and run predictions on it in the cloud.

Starting today, developers can begin building apps with the DALL·E API.

Weaviate 1.16 introduces New Filter Operators, Distributed Backups, Centroid Module, Node Status API, Azure-based OIDC, and more. Lear all about it.

New users can start creating straight away. Lessons learned from deployment and improvements to our safety systems make wider availability possible.

Running ML Model Inference in production is hard. You can use Weaviate – a vector database – with Hugging Face Inference module to delegate the heavy lifting.

Neural networks are well known to be over-parameterized and can often easily fit data with near-zero training loss with decent generalization performance on tes

How to run Stable Diffusion locally so you can hack on it

Self-Supervised Retrieval can surpass BM25 and Supervised techniques. This technique also pairs very well alongside BM25 in Hybrid Retrieval. Learn more about i

How to use Replicate to integrate Stable Diffusion into hacks, apps, and projects

A tutorial for building a chat bot that replies to prompts with the output of a text-to-image model.

We are improving our AI systems’ ability to learn from human feedback and to assist humans at evaluating AI. Our goal is to build a sufficiently aligned AI syst