Researchers Test GPT-6 Astra Across Robotics and Biomedicine
In an interview republished by 36Kr and TMTPost, researchers who tested GPT-6 Astra reported strong adaptation, iterative improvement, Real-to-Sim and computer-use capabilities. They also identified practical limits, including slower responses, hallucinations during overnight tasks, an inability to perform dual-arm tasks such as folding clothes, and a need for smaller robot-specific models. A biomedical researcher said his team now uses Astra for internal biomedical work, particularly operating older scientific software and handling workflows lasting 8 to 24 hours.
The interview offers concrete first-hand observations about where GPT-6 Astra appears useful in robotics and biomedicine, alongside unresolved questions about speed, reliability, deployment size and dexterity.
回顾9月,AI圈的重磅新闻一个接一个。
9月4日,OpenAI发布GPT-6 Astra。这个发布报告里只字未提机器人的通用模型,转身就空降了机器人评测榜首,甚至让一批机器人Benchmark直接失效。有人半开玩笑地说:机器人领域等了这么多年的“GPT时刻”,居然是GPT自己。
Meta Muse给每个用户配了一台云端电脑,两周就冲上美区App Store榜首;竞争对手Instinct一个月内估值从25亿涨到100亿美元。个人Agent掀起了新一轮烧钱获客大战,可它们能打开的网页,正在一个一个被关上。
AI for Science方面,Claude只用了21小时就找到一种结构像CRISPR的新酶系统ART,AI圈很兴奋,因为AI开始做科学发现了,生物圈的反应则相对平淡,大家想在实验室里验证了再说,为什么会有这种反差现象?
#GPT-6 Astra#OpenAI#Claude#Meta Muse#Instinct#MultiQC#Pokee.ai#Unitree Robotics