Benchmark tests whether AI coding agents can solve research problems

Original: RT @IntologyAI: Can coding agents do research?

Source: x.com ↗