SuperCLUE General (July 2025) - Hallucination Control: leaderboard
SuperCLUE general leaderboard, July 2025 edition (2025年7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. 47 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4 (Thinking) | 90.24 | #170 (Claude Opus 4) |
| 2 | Doubao-Seed-1.6-thinking-250715 | 89.59 | #169 |
| 3 | Gemini 2.5 Flash | 89.36 | #255 |
| 4 | Hunyuan-T1-20250711 | 83.39 | #277 |
| 5 | GPT-5 (High) | 83.08 | #105 (GPT-5) |
| 6 | Step 2 16K | 83.03 | #442 |
| 7 | Gemini 2.5 Pro | 83 | #159 |
| 8 | DeepSeek R1 0528 | 82.28 | #232 |
| 9 | DeepSeek V3.1 (Thinking) | 82.19 | #275 (DeepSeek V3.1) |
| 10 | DeepSeek V3 (0324) | 81.71 | #349 |
| 11 | Grok 4 | 79.65 | #189 |
| 12 | GLM-4.5 | 78.42 | #283 |
| 13 | O3 (High) | 77.41 | #132 (O3) |
| 14 | O4 Mini (High) | 75.9 | #185 (O4 Mini) |
| 15 | SenseNova V6 Reasoner | 75.57 | #317 |
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-july-2025-hallucination-control · How It Works · Data refreshed daily, snapshot 2026-11-05.