GlobeBench measures LLM text world-model simulation accuracy
Across 1,632 tasks, GPT-6 Astra leads 20 models at 49.57%, and 631 tasks stay unsolved.
Stories per day · 30 days
1 stories, newest first · one card per story
Stories tagged #globebench
1 stories, newest first · one card per story
Across 1,632 tasks, GPT-6 Astra leads 20 models at 49.57%, and 631 tasks stay unsolved.
Today 1.6K readers · 2.4K views · 774 reads · all time 189K views