Skip to main content
YZ Index

Winzheng · Engineering Judgment Rankings

Architecture review, risk assessment, engineering decision-making.

Overall Rankings Code Execution Rankings Grounding Rankings Engineering Judgment Rankings Task Communication Rankings Value Rankings Changes
Side Dimension Rankings — Communication + Judgment sub-dimension performance
# Model Engineering Judgment Code Execution Overall
🥇 GPT-5.5 gpt
90
80.1 71.3
🥈 GPT-6 Astra gpt
90
84.4 75.2
🥉 GPT-6.1 Sol gpt
89.6
93 81.8
4 Doubao Pro doubao
87.8
87.4 76.5
5 GPT-6 Sol gpt
87.5
85.6 74.2
6 GPT-o3 gpt
84.6
65 68.1
7 GPT-6 Luna gpt
83.2
84.8 76.3
8 Gemini 3.1 Pro gemini
82.8
52.4 60.1
9 DeepSeek V4 Pro DeepSeek
81.7
62.1 65
10 Claude Opus 4.7 claude
78.1
76.9 79.8
11 Claude Sonnet 4.6 claude
70.3
76.1 72.2
12 Gemini 2.5 Pro gemini
69.3
52.6 57.1
13 Grok 4 grok
52.1
69.8 65.8
14 GLM-4.6 zhipu
20
31.8 48.7
15 Qwen3 Max qwen Data backfilling — — —

Run #397 · Formula v7 · Judge v6.5 · Benchmark v7

Full evaluation: 128 questions from pool of 100yz.rankings.footer.full_eval_suffix · Latest Evaluation2026-10-26 05:06 SGT · Models15models

Winzheng YZ Index — Winzheng · YZ Index