SuperCLUE-VLM - Overall: leaderboard
SuperCLUE-VLM evaluates vision-language models on visual understanding, cross-modal reasoning and multimodal dialogue. Headline board of the newest workbook the superclueai.com vertical lists; score 0-100, higher is better.
Metric: Score. Source: www.superclueai.com. Saturation forecast: Around February 2027. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 83.9 | #6 |
| 2 | Gemini 3.8 Flash | 79.15 | #30 |
| 3 | Qwen 3.8 Max (0902) | 76.45 | #36 |
| 4 | Claude Fable 5.1 | 75.13 | #10 |
| 5 | Seed 2.1 Pro | 73.84 | #73 |
| 6 | Kimi K3 | 73.8 | #35 |
| 7 | Muse Spark 1.3 | 71.8 | #31 |
| 8 | GLM-5.3 Flash | 69.94 | #77 |
| 9 | Grok 4.6 | 63.55 | #29 |
| 10 | Step 3.7 Flash | 60.41 | #247 |
| 11 | DeepSeek V4.1 Flash | 56.49 | #37 |
| 12 | MiMo-V2.5 | 55.57 | #196 |
| 13 | MiniMax-M3 | 50.22 | #151 |
| 14 | MiniCPM-V-4.6 | 25.39 | #1014 |
No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=superclue-vlm-overall · How It Works · Data refreshed daily, snapshot 2026-11-05.