List your company

Instruction Following on ComplexBench (test)

79.8CSRCRGC
59.5264.78570.0575.315Jun 2, 2026

Evaluation results

MethodLinks
CRGCBase Model=Claude-3.7-Sonnet, Model Type=Closed-source2026.0679.883.681
CRGCBase Model=GPT-4o, Model Type=Closed-source2026.0678.682.579.8
CRGCBase Model=Gemini-2.5-Pro, Model Type=Closed-source2026.0677.581.478.9
CRGCBase Model=Llama-3.1-72B, Model Type=Open-source2026.0675.679.476.8
CRGCBase Model=Qwen2.5-72B, Model Type=Open-source2026.0674.878.776.1
Self-Reflection (SR)Base Model=Claude-3.7-Sonnet, Model Type=Closed-source2026.0674.17875.5
Self-Reflection (SR)Base Model=GPT-4o, Model Type=Closed-source2026.0673.277.174.6
Chain-of-Thoughts (CoT)Base Model=Claude-3.7-Sonnet, Model Type=Closed-source2026.0672.376.273.7
Self-Reflection (SR)Base Model=Gemini-2.5-Pro, Model Type=Closed-source2026.0672.376.373.8
CRGCBase Model=Mixtral-8x22B, Model Type=Open-source2026.0672.276.173.5
Chain-of-Thoughts (CoT)Base Model=GPT-4o, Model Type=Closed-source2026.0671.575.472.8
Chain-of-Thoughts (CoT)Base Model=Gemini-2.5-Pro, Model Type=Closed-source2026.0670.674.572
Self-Reflection (SR)Base Model=Llama-3.1-72B, Model Type=Open-source2026.0669.773.871
Self-Reflection (SR)Base Model=Qwen2.5-72B, Model Type=Open-source2026.0668.97370.3
Chain-of-Thoughts (CoT)Base Model=Llama-3.1-72B, Model Type=Open-source2026.0667.871.969.1
Standard Prompting (SP)Base Model=Claude-3.7-Sonnet, Model Type=Closed-source2026.0667.471.668.8
Chain-of-Thoughts (CoT)Base Model=Qwen2.5-72B, Model Type=Open-source2026.0666.870.968.2
Standard Prompting (SP)Base Model=GPT-4o, Model Type=Closed-source2026.0666.770.868.1
Self-Reflection (SR)Base Model=Mixtral-8x22B, Model Type=Open-source2026.0666.470.567.7
Standard Prompting (SP)Base Model=Gemini-2.5-Pro, Model Type=Closed-source2026.0665.87067.2
Chain-of-Thoughts (CoT)Base Model=Mixtral-8x22B, Model Type=Open-source2026.0664.568.665.8
Standard Prompting (SP)Base Model=Llama-3.1-72B, Model Type=Open-source2026.0663.467.764.7
Standard Prompting (SP)Base Model=Qwen2.5-72B, Model Type=Open-source2026.0662.566.963.9
Standard Prompting (SP)Base Model=Mixtral-8x22B, Model Type=Open-source2026.0660.364.561.7