ResearchAug 28, 2026 · 03:552 sources

Claude Opus 5 Solves Only 30% of New Scientific AI Benchmark

From huggingnews.com

AI;DR

A Stanford-led community effort released Terminal-Bench-Science v0.1 to evaluate AI agents on research workflows across scientific domains. The initial benchmark contains 70 tasks, and Claude Opus 5 solved approximately 30% of them.

Read the sourceSign in to suggest

Sources

+ 1 more
  1. SUPPORT@madiator“it will spur progress in science!”x.com

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today