Skip to main content

Share story

AI

ARC Prize reports wide GPT-6 Astra score gap across test harnesses

ARC Prize reports wide GPT-6 Astra score gap across test harnesses Image: Primary
ARC Prize published results showing GPT-6 Astra scored 62.7% on ARC-AGI-3 under its standard harness and 99.9% using OpenAI's Provider Adapter. The benchmark organization said the adapter preserves opaque reasoning state between requests and compacts longer conversations, unlike its minimal standard interface. ARC Prize said it is not claiming the result establishes AGI and will publish both harness results side by side. The source also reports that the compared 99.9% Astra and 7.8% Sol figures came from different harnesses.
Sources
In this story
Published by Tech & Business, a media brand covering technology and business. This story was sourced from The Next Web and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
AI Capital
AI Capital

Arda raises $230M at $1.5B valuation for manufacturing AI

Arda raised $230M at a $1.5B valuation, CNBC reported. The startup develops AI software and agents to accelerate assembly line programming and manufacturing automation. OpenAI's former research chief has been building the company ...

AI Security
AI Security

GitLab opens beta dashboard for auditing coding agents

GitLab introduced a beta AI Governance Dashboard for Premium and Ultimate customers that brings its own coding agents and connected external agents into one audit view. It links agent activity to developers and projects, showing s...