SuperCLUE General (November 2026) - Hallucination Control: leaderboard
SuperCLUE general leaderboard, November 2026 edition: hallucination-control dimension (幻觉控制), measuring control of unsupported answers in the Chinese-language general evaluation. Native score, higher is better; kept separate from other dimensions and monthly editions. Source: https://www.superclueai.com/generalpage
Metric: Score. Source: www.superclueai.com. Saturation forecast: Around December 2027. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon | 83.19 | #11 |
| 2 | Kimi K3.1 Pro | 67.78 | #28 |
| 3 | GPT-6.1 Sol | 63.61 | #10 |
| 4 | Grok 4.7 | 56.95 | #43 |
| 5 | GLM-5.3 Flash | 48.89 | #119 |
| 6 | DeepSeek V4.1 Flash (1017) | 43.89 | #38 |
| 7 | GLM-5.3 | 43.61 | #62 |
| 8 | MiniMax-M3 | 39.86 | #265 |
| 9 | Step 5 Preview | 38.89 | #105 |
| 10 | Hy4 preview | 33.34 | #66 |
| 11 | Qwen 3.8 Max (0902) | 33.05 | #49 |
| 12 | Qwen 3.8 Flash | 32.36 | #109 |
| 13 | MiniMax-M3.1-Flash | 31.11 | #104 |
| 14 | MiMo-V2.6-Flash | 27.78 | #150 |
| 15 | MiMo-V2.6-Pro | 25.28 | #68 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-november-2026-hallucination-control · How It Works · Data refreshed daily, snapshot 2026-11-12.