Report
RSIGym launches as an open-source research environment for AI agents
Evolvent AI says Opus 5 led its RSI-Index study of six research agents across five benchmarks, scoring 0.4809.
TLDR
Evolvent AI says RSIGym gives AI research agents reusable services to experiment with training data, model settings and execution harnesses. The team released its code, experiment configurations and evaluation logs. Its RSI-Index measures the mean share of the remaining performance gap closed across five benchmarks; the team reports that Opus 5 led six research agents with a score of 0.4809 under a $500 platform-service budget per benchmark run.
Combined views
27
1 Source, first seen ago
