Share your thoughts, 1 month free Claude Pro on usSee more

Pedagogical Dialogue Evaluation on SHAPE (test)

93.66Safety Score

Claude Opus 4.5

Updated 1mo ago

Evaluation Results

Method	Links
Claude Opus 4.5 2026.04		93.66	100	74.44
Claude Haiku 4.5 2026.04		93.66	100	76.69
Gemini 2.5 Pro 2026.04		93.66	100	72.18
Gemini 2.5 Flash 2026.04		93.66	100	82.71
Gemini 2.5 Flash-Lite 2026.04		93.66	100	78.2
Claude Haiku 4.5 2026.04		92.96	100	74.24
Qwen3-80B 2026.04		92.96	100	74.24
Claude Opus 4.5 2026.04		92.25	100	68.7
Gemini 2.5 Pro 2026.04		92.25	100	71.76
Gemini 2.5 Flash 2026.04		92.25	100	75.57
Qwen3-80B 2026.04		92.25	100	70.99
Claude Opus 4.5 2026.04		91.55	100	73.08
Gemini 2.5 Flash-Lite 2026.04		91.55	100	83.85
Gemini 2.5 Flash-Lite 2026.04		90.85	100	83.72
Qwen3-80B 2026.04		90.85	100	65.89
GPT-5 mini 2026.04		90.14	100	93.75
Claude Haiku 4.5 2026.04		89.44	100	66.93
GPT-5 2026.04		89.44	100	94.49
GPT-5 mini 2026.04		88.46	100	89.13
GPT-5 2026.04		86.62	100	94.31
GPT-5 mini 2026.04		86.62	100	93.5
Qwen3-32B 2026.04		86.62	100	78.86
Qwen3-8B 2026.04		86.62	100	74.8
GPT-5 2026.04		85.92	100	95.08
GPT-5 nano 2026.04		79.58	100	86.73
Gemini 2.5 Pro 2026.04		77.46	98.28	75.45
GPT-5 nano 2026.04		77.46	100	90.91
GPT-5 nano 2026.04		73.94	100	84.76
Gemini 2.5 Flash 2026.04		39.44	98.28	71.43
Qwen3-32B 2026.04		15.49	100	63.64
Qwen3-32B 2026.04		7.04	100	50
Qwen3-8B 2026.04		3.52	100	40
Qwen3-8B 2026.04		1.41	100	0