Yhn TechnologyRobotics first seen 3 h ago, last 56 min ago, peak #11
Frontier AI models score under 2/100 at controlling simulated humanoid
Original: Frontier models score under 2/100 controlling a simulated humanoid
A new robotics benchmark from Vals AI finds that today's frontier language models score under 2 out of 100 when tasked with controlling a simulated humanoid robot. The result highlights how far current AI remains from general-purpose physical control, despite strong performance on text and reasoning tasks, and is drawing attention in the AI and robotics communities.
Why now: It starkly quantifies the gap between frontier models' reasoning abilities and their ability to control robots in physical-style environments.
Vals AIfrontier AI modelshumanoid robots
Rank over time, top of the chart is #1. 3 snapshots from 3 h ago to 56 min ago.
Evidence
- Frontier models score under 2/100 controlling a simulated humanoid · miyukimatthews · 7
API: https://socialmediatrends-api.osmike.com/v1/trends/2305561