danielxu0208/rsi_eval is a lightweight Python SDK and CLI that attempts to evaluate the recursive self-improvement (RSI) technique across multiple agent benchmarks through a single interface.
danielxu0208/rsi_eval is a lightweight Python SDK and CLI that attempts to evaluate the recursive self-improvement (RSI) technique across multiple agent benchmarks through a single interface.