D2VBench - Social Order: leaderboard

Metric: Value-alignment score (0-100) on the social-order scenarios; 10,000 everyday value-dilemma scenarios answered free-form; GLM-4.6 and Qwen3-Max map each response to the matching options and score its coverage of annotated interpretability points over five dimensions (0-100), the two judges averaged. Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked.

Top models

#ModelScore
1GPT-5.165.96
2Gemini 3 Pro (Preview)64.63
3DeepSeek R1 052862.58
4GLM-4.661.03
5MiniMax-M259.6
6Claude Haiku 4.5 (20251001)57.04
7Kimi K2 090556.99
8Seed-1.655.09

Interactive version: theaggregate.ai/benchmark?slug=d2vbench-social-order · How It Works · Data refreshed daily, snapshot 2026-09-29.