MME-CC: leaderboard
MME-CC — multimodal evaluation of cognitive-capability tasks over images; overall score.
Metric: Overall Score (%). Source: randomtutu.github.io. Saturation forecast: Around January 2027. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 42.66 | #152 |
| 2 | GPT-5 (High) | 40.25 | #98 (GPT-5) |
| 3 | Gemini 2.5 Flash | 37.57 | #245 |
| 4 | O4 Mini (High) | 35 | #180 (O4 Mini) |
| 5 | GPT-4.1 | 32.14 | #248 |
| 6 | GLM-4.5V | 30.45 | #346 |
| 7 | GPT-4o (2024-11-20) | 26.88 | #381 |
| 8 | Qwen 2.5 VL 72B Instruct | 23.59 | #377 |
| 9 | GLM-4.1V-9B (Thinking) | 19.3 | #472 (GLM-4.1V-9B) |
| 10 | Qwen 2.5 VL 32B Instruct | 14.39 | #453 |
| 11 | InternVL3-8B | 11.36 | #620 |
| 12 | Qwen 2.5 VL 7B Instruct | 7.5 | #654 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mme-cc · How It Works · Data refreshed daily, snapshot 2026-10-20.