Running thousands of LLMs on one GPU is now possible with S-LoRA

Researchers from Stanford University and UC Berkeley have developed S-LoRA, a technique that significantly reduces the…

iTransformer: Rethinking Transformer Architecture for Enhanced Time Series Forecasting

The Transformer model, successful in natural language processing and computer vision, is now emerging in time…

Role play with large language models – Nature

Language model-based dialogue agents can’t literally lie or believe falsehoods, but they can role-play characters that…

GPT-4 Vision Chrome Extension

Explore the new GPT-4 Vision Chrome extension, a proof-of-concept tool designed to streamline web-based tasks. With…

Boosting Code LLMs Through Innovative Multitask Fine-Tuning

A new study proposes an innovative approach to enhancing the capabilities of Code LLMs through multi-task…

10 Major AI Updates at GitHub Universe 2023

Microsoft’s GitHub Copilot software has seen a 40% rise in paying customers in Q3, with over…

Google AI Researchers Found Something Their Bosses Might Not Be Happy About

Google DeepMind researchers have found that current AI models, specifically transformer models like OpenAI’s GPT-2, struggle…

Setting time on fire and the temptation of The Button

The integration of AI into everyday tools like Google Docs and Microsoft Office is set to…

Chinese scientists’ attack on ChatGPT shows how criminals can exploit AI

Researchers in Beijing have found vulnerabilities in commercial AI models, including ChatGPT, that could allow them…

China Says It Will Roll Out Humanoid Robots by 2025

China plans to produce its first humanoid robots by 2025, aiming to alleviate human workload in…

Offensive and Defensive AI: Let’s Chat(GPT) About It

ChatGPT, a popular AI chatbot, can be exploited by cybercriminals for data exfiltration, spreading misinformation, and…

Everything announced at OpenAI’s first developer event

OpenAI unveiled several new products at its first developer event, including the GPT-4 Turbo model for…

OpenAI announces new models and developer products at DevDay

OpenAI has launched the Assistants API, a tool designed to help developers create AI-powered applications. The…

DeepSeek-Coder: Let the Code Write Itself

DeepSeek Coder, a series of code language models, offers state-of-the-art performance in coding capabilities. Trained on…

Making Whisper Models Faster and Smaller Through Knowledge Distillation

Recent advances in self-supervised pre-training have led to impressive gains in speech recognition performance. Models like…

Phind’s New Model Matches GPT-4 in Coding at 5x the Speed

A company Phind has unveiled a new model that achieves coding abilities on par with OpenAI’s…

Yarn-Mistral-7b-128k

The Nous-Yarn-Mistral-7b-128k is a cutting-edge language model designed for long context. It’s an extension of Mistral-7B-v0.1,…

YaRN: Efficient Context Window Extension of Large Language Models

The YaRN context window extension method enhances the efficiency of large language models. The team has…

Telling GPT-4 you’re scared or under pressure improves performance

AI models like GPT-4 perform better when users express emotions such as urgency or stress, according…

Video Game Created Entirely With ChatGPT, DALL-E 3, and Midjourney

Javi Lopez has created a game, “Angry Pumpkins,” using AI tools like ChatGPT and Midjourney, demonstrating…