List your company

Segment Intrusion on Segment Intrusion Restaurant domain 1.0 (test)

98SI-EHuman Performance
24.232843.383962.53581.6861May 18, 2026

Evaluation results

MethodLinks
Human Performance2026.059897978794.75
o4-mini2026.059883.719577.1888.47
o3-mini2026.0597.582.129474.6387.06
gemini-2.5-flash-preview-04-172026.0597.582.576.2763.9880.06
claude-3.7-sonnet-202502192026.0596.7677.59572.9985.56
llama-3.3-70b-instruct-turbo2026.0595.5274.1988.7864.8980.84
deepseek-v32026.059578.585.5770.2282.32
gpt-4o2026.0593.579.3189.2873.4883.89
qwen2.5-72b-instruct-turbo2026.0591.277491.0465.8680.54
claude-3.5-haiku-202410222026.0590.5172.558266.3477.85
gemini-2.0-flash2026.059066.586.2863.7776.64
gemini-2.0-flash-lite2026.0579.8262.5776.2466.6771.32
qwen2.5-7b-instruct-turbo2026.0574.8153.540.131.7550.04
llama-4-maverick-17b-128e-instruct-fp8Precision=fp82026.0555.9445.5921.4819.8235.71
llama-3.2-3b-instruct-turbo2026.0527.0720.7610.497.516.45