SuperCLUE-VLM (April 2026) - Overall: leaderboard
SuperCLUE-VLM evaluates vision-language models in Chinese on basic visual cognition, visual reasoning and applied visual tasks. April 2026 edition (2026年4月): headline 总分 of the 总榜 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. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Doubao-Seed-2.0-Pro-260215 | 90.66 | #86 |
| 2 | Gemini 3.1 Pro (Preview) | 89.35 | #64 |
| 3 | Qwen 3.5 397B A17B (Thinking) | 88.5 | #154 (Qwen 3.5 397B A17B) |
| 4 | Qwen 3.6 Plus (Thinking) | 86.73 | #124 (Qwen 3.6 Plus) |
| 5 | Kimi K2.6 (Thinking) | 84.65 | #113 (Kimi K2.6) |
| 6 | Qwen 3.6 35B A3B (Thinking) | 84.57 | #238 (Qwen 3.6 35B A3B) |
| 7 | GLM-5V Turbo | 84.18 | #172 |
| 8 | Kimi K2.5 (Thinking) | 83.87 | #152 (Kimi K2.5) |
| 9 | GPT-5.4 (xHigh) | 80.79 | #83 (GPT-5.4) |
| 10 | Grok 4.20 0309 (Reasoning) | 78.01 | #120 (Grok 4.20 0309) |
| 11 | Step-3 | 73.53 | #380 |
| 12 | Mistral Small 4 | 42.13 | #522 |
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-april-2026-overall · How It Works · Data refreshed daily, snapshot 2026-11-05.