List your company

Multi-turn instruction-following on Multi-IF

95.02Turn 1 ScoreGemini 3.1 Pro
69.519276.139682.7689.3804May 21, 2026

Evaluation results

MethodLinks
Gemini 3.1 ProThinking mode=true2026.0595.0289.5784.89
GPT-5.5Thinking mode=true2026.0594.989.3384.46
GLM5.1Thinking mode=true2026.0593.9486.4782.77
Kimi K2.6Thinking mode=true2026.0593.8787.1582.29
DeepSeek-V4-ProThinking mode=true2026.0593.6287.0281.69
GPT-5.5Thinking mode=false2026.0593.2587.2182.02
GLM5.1Thinking mode=false2026.0592.2485.2579.67
Qwen3.5-397B-A17BThinking mode=true2026.0590.8184.1779.56
Kimi K2.6Thinking mode=false2026.0590.684.9377.48
Qwen3.5-397B-A17BThinking mode=false2026.0590.581.6275.86
DeepSeek-V4-ProThinking mode=false2026.0590.3683.4576.6
Hy-MT2-30B-A3BThinking mode=false2026.0590.172.7366.66
Gemma4-26B-A4BThinking mode=false2026.0589.1781.8975.29
Gemma4-E4BThinking mode=false2026.0586.8178.0471.73
Qwen3.6-35B-A3BThinking mode=false2026.0584.8478.1271.48
Gemma4-E2BThinking mode=false2026.0580.7970.5963.81
Hy-MT2-7BThinking mode=false2026.0579.5366.554.35
Hy-MT2-1.8BThinking mode=false2026.0570.549.7935.75