Computer use tool

I’ve been using large language models (LLMs) for a while now. They accelerate and improve my output at work, to the point that losing access to them would make me feel slightly impaired in some areas. Rather than fearing WriterBot, I’m embracing the additional capabilities it grants. At the same time, I’m extremely conscious of their limitations, which are abundant. Let me tell you how LLMs are helping me in my everyday work as a documentation engineer and where they’re unable to assist me.

Two big announcements from Anthropic today: a new Claude 3.5 Sonnet model and a new API mode that they are calling computer use. (They also pre-announced 3.5 Haiku, but that’s …

A refreshed, more powerful Claude 3.5 Sonnet, Claude 3.5 Haiku, and a new experimental AI capability: computer use.

Building agentic malware-analysis workflows that combine strong initial triage with deeper automated recovery of malware structure and behavior.

Experiments with Google Gemini's Vision capabilities for LLM driven/aided web navigation and desktop manipulation - byt3bl33d3r/gemini-web-navigator

A few days ago, Anthropic released Claude Computer Use, which is a model + code that allows Claude to control a computer. It takes screenshots to make …

Zhengdong Wang’s personal website

Most of the world's most cutting-edge AI systems are made using surprisingly similar methods. Here are the basics.

thoughts on building agentic AI in the reasoning era

A refreshed, more powerful Claude 3.5 Sonnet, Claude 3.5 Haiku, and a new experimental AI capability: computer use.