YZ Index
YZ Index · Integrity Rating
Gateway mechanism: Models must pass integrity checks to be ranked.
Gemini 2.5 Pro
gemini
PASS
Integrity Score 93.3
recommended
Claude Opus 4.7
claude
PASS
Integrity Score 90
recommended
Claude Sonnet 4.6
claude
PASS
Integrity Score 86.7
recommended
豆包 Pro
doubao
PASS
Integrity Score 86.7
recommended
GPT-6 Astra
gpt
PASS
Integrity Score 86.7
recommended
GPT-6 Sol
gpt
PASS
Integrity Score 86.7
recommended
GPT-6.1 Sol
gpt
PASS
Integrity Score 86.7
recommended
Gemini 3.1 Pro
gemini
PASS
Integrity Score 83.3
recommended
DeepSeek V4 Pro
DeepSeek
PASS
Integrity Score 81.7
recommended
GPT-5.5
gpt
PASS
Integrity Score 81.7
recommended
GPT-6 Luna
gpt
PASS
Integrity Score 78.3
recommended
GPT-o3
gpt
PASS
Integrity Score 78.3
recommended
Grok 4
grok
PASS
Integrity Score 78.3
recommended
GLM-4.6
zhipu
WARN
Integrity Score 50
neutral
Qwen3 Max
qwen
FAIL
Integrity Score —
—
Overall score capped at 74
Methodology
Integrity Rating is based on 25 tasks (including 12 honesty_under_pressure stress tests), assessing whether models honestly acknowledge their own errors without deflecting or downplaying. >= 60 points: pass, 40-59: warn, < 40: fail. Detailed Methodology →
Integrity Rating is based on 25 tasks (including 12 honesty_under_pressure stress tests), assessing whether models honestly acknowledge their own errors without deflecting or downplaying. >= 60 points: pass, 40-59: warn, < 40: fail. Detailed Methodology →