SuperCLUE General (September 2025) - Hallucination Control: leaderboard

SuperCLUE general leaderboard, September 2025 edition (2025年9月): hallucination control (幻觉控制) score (总排行榜 sheet). SuperCLUE's periodic Chinese-language general evaluation uses a new question set each edition, so editions are separate boards. Score 0-100, higher is better.

Metric: Score. Source: www.superclueai.com. Saturation forecast: Estimated already saturated. 37 models tracked.

Top models

#ModelScoreOverall rank
1Claude Opus 4.1 (Thinking)85.23#146 (Claude Opus 4.1)
2Gemini 3 Pro (Preview)81.51#71
3Claude Sonnet 4.5 (Thinking)78.61#148 (Claude Sonnet 4.5)
4GPT-5 (High)78.34#105 (GPT-5)
5Hunyuan-T1-2025082277.96#306
6Doubao-Seed-1.6-thinking-25071577.78#169
7O3 (High)77.47#132 (O3)
8GPT-5.1 (High)77.11#145 (GPT-5.1)
9Grok 473.52#189
10Gemini 2.5 Pro73.17#159
11GLM-4.673.08#265
12Gemini 2.5 Flash72.36#255
13Kimi K2 (Thinking)72.23#252 (Kimi K2)
14DeepSeek V3.1 Terminus71.44#224
15DeepSeek V3.2 Exp (Thinking)71.26#241 (DeepSeek V3.2 Exp)

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=superclue-general-september-2025-hallucination-control · How It Works · Data refreshed daily, snapshot 2026-11-05.