SuperCLUE General (2025 Annual) - Hallucination Control: leaderboard
SuperCLUE general leaderboard, 2025 Annual edition (2025年度测评): hallucination control (幻觉控制) score from the 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: Around February 2027. 34 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 (High) | 88.56 | #115 (GPT-5.2) |
| 2 | Claude Opus 4.5 (Thinking) | 88.31 | #87 (Claude Opus 4.5) |
| 3 | Claude Opus 4.6 (High) | 85.87 | #67 (Claude Opus 4.6) |
| 4 | Gemini 3.1 Pro (Preview) (High) | 84.42 | #64 (Gemini 3.1 Pro (Preview)) |
| 5 | GLM-4.7 | 83.85 | #202 |
| 6 | Gemini 3 Pro (Preview) | 83.16 | #71 |
| 7 | Qwen 3.5 397B A17B (Thinking) | 83.13 | #154 (Qwen 3.5 397B A17B) |
| 8 | Doubao-Seed-2.0-Pro-260215 | 82.39 | #86 |
| 9 | Gemini 3 Flash (Preview) | 81.57 | #90 |
| 10 | Qwen 3.5 27B (Thinking) | 81.14 | #192 (Qwen 3.5 27B) |
| 11 | GLM-5 | 81.05 | #149 |
| 12 | Grok 4 | 78.9 | #189 |
| 13 | Kimi K2.5 (Thinking) | 78.54 | #152 (Kimi K2.5) |
| 14 | DeepSeek V3.2 (Thinking) | 76.57 | #210 (DeepSeek V3.2) |
| 15 | Qwen 3.5 122B A10B (Thinking) | 74.72 | #184 (Qwen 3.5 122B A10B) |
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-2025-annual-hallucination-control · How It Works · Data refreshed daily, snapshot 2026-11-05.