List your company

Instruction Following on IFBench In-Distribution (F1 Metrics)

41.1Initial F1AutoPyVerifier
25.08429.24233.437.558Apr 24, 2026

Evaluation results

MethodLinks
AutoPyVerifierVerifier Model=Gemini-3.1 Fl2026.0441.151.710.6
AutoPyVerifierVerifier Model=Claude Hai-4.52026.0439.559.119.6
AutoPyVerifierVerifier Model=Gemini-3.1 Pro2026.0429.481.552.1
AutoPyVerifierVerifier Model=GPT-5.42026.0428.542.614.1
AutoPyVerifierVerifier Model=GPT-5.4 mini2026.0425.739.513.8