List your company

Instruction Following on IFBench Out-of-Distribution (F1 Metrics)

41.5Initial F1AutoPyVerifier
21.63626.79331.9537.107Apr 24, 2026

Evaluation results

MethodLinks
AutoPyVerifierVerifier Model=Claude Hai-4.52026.0441.563.121.6
AutoPyVerifierVerifier Model=Gemini-3.1 Fl2026.0433.541.37.8
AutoPyVerifierVerifier Model=GPT-5.42026.0425.137.112
AutoPyVerifierVerifier Model=Gemini-3.1 Pro2026.0425.179.554.4
AutoPyVerifierVerifier Model=GPT-5.4 mini2026.0422.441.519.1