SuperCLUE-VLM (February 2026) - Overall: leaderboard
SuperCLUE-VLM evaluates vision-language models in Chinese on basic visual cognition, visual reasoning and applied visual tasks. February 2026 edition (2026年2月): headline 总分 of the 总榜成绩 sheet. Each dated workbook is a separate evaluation round, kept apart from the current board. Score 0-100, higher is better.
Metric: Score. Source: www.superclueai.com. Saturation forecast: Around February 2027. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 84.87 | #64 |
| 2 | Doubao-Seed-2.0-Pro-260215 | 81.2 | #86 |
| 3 | Qwen 3.5 Plus (Thinking) | 80.24 | #133 (Qwen 3.5 Plus) |
| 4 | Kimi K2.5 (Thinking) | 79.95 | #152 (Kimi K2.5) |
| 5 | Qwen 3.5 397B A17B (Thinking) | 79.35 | #154 (Qwen 3.5 397B A17B) |
| 6 | ERNIE 5.0 | 73.77 | #243 |
| 7 | GPT-5.2 (High) | 73.31 | #115 (GPT-5.2) |
| 8 | Claude Opus 4.6 (High) | 72.99 | #67 (Claude Opus 4.6) |
| 9 | Step-3 | 66.62 | #380 |
| 10 | GLM-4.6V | 65.76 | #330 |
| 11 | Grok 4.1 Fast (Reasoning) | 51.05 | #227 (Grok 4.1 Fast) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=superclue-vlm-february-2026-overall · How It Works · Data refreshed daily, snapshot 2026-11-05.