AI repos to watch
News from Alibaba and Andrej Karpathy in the AI space
Qwen
Qwen is an open-source large language model proposed by Alibaba Cloud. It includes various models such as Qwen-1.8B, Qwen-7B, Qwen-14B, and Qwen-72B, each designed for unique use cases. These models are trained on a large amount of data and are intended for tasks such as code execution, language understanding, and multimodal applications.
Qwen’s open-source nature allows developers, researchers, and organizations to access and utilize these models for a wide range of applications, including natural language processing, image understanding, and audio interpretation.
The release of Qwen reflects Alibaba Cloud’s efforts to contribute to the advancement of AI technology through open collaboration and innovation.
Minbpe
Built by and recently released by Andrej Karpathy. Andrej Karpathy has recently left OpenAI to pursue other avenues of research. He is widely considered a significant thought leader in the AI space. This means his work in open source LLMs and other AI tools is a must watch for those following AI developments.
BPE stands for Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.
I cover more AI tools and news in my newsletter BrainScriblr. I give topics more depth there. So why not subscribe as it is free.
GPTscript
GPTScript, which is designed to automate interactions with Large Language Models (LLMs), specifically OpenAI’s models. GPTScript aims to provide a natural language-based programming experience, allowing users to use largely natural language syntax for tasks such as code execution, language understanding, and multimodal applications.
The language is intended to be easy to learn and use, enabling a wide range of functionalities, including interacting with AI models, running traditional scripts, and making external HTTP service calls. The development of GPTScript represents an effort to enhance the usability and accessibility of large language models for various applications.
