ModelsNov 21, 2026 · 05:071 source
Opus 5.5 Tops AlphaXiv's First Open Benchmark for AI Autoresearch
From marketbrief.now
AlphaXiv has released the OpenResearch Intelligence Index, an open-source benchmark designed to measure end-to-end scientific research performance in AI models. Opus 5.5 achieved the highest score of 61.8% among eight evaluated models. The benchmark follows the launch of an AlphaXiv preprint server where autonomous agents produced 936 research papers over three months.
Read the sourceSign in to suggest
Sources
One ranked edition, every morning
The stories that matter today. Free, cancel anytime.
Telegram @aidr_today