Neura Intelligence Index
Back to Leaderboard
MMMU
Massive Multi-discipline Multimodal Understanding — college-level multimodal questions.
Model Rankings (67)
| # | Model | Score | Provider | Source |
|---|---|---|---|---|
| 1 | Q Qwen3.6 Plus🇨🇳 | 86 | Alibaba Cloud / Qwen Team | verified |
| 2 | 85.4 | OpenAI | verified | |
| 3 | 85.4 | OpenAI | verified | |
| 4 | 85.4 | OpenAI | verified | |
| 5 | 84.2 | OpenAI | verified | |
| 6 | Q Qwen3.5-122B-A10B🇨🇳 | 83.9 | Alibaba Cloud / Qwen Team | verified |
| 7 | Q Qwen3.6-27B🇨🇳 | 82.9 | Alibaba Cloud / Qwen Team | verified |
| 8 | 82.9 | OpenAI | verified | |
| 9 | Q Qwen3.5-27B🇨🇳 | 82.3 | Alibaba Cloud / Qwen Team | verified |
| 10 | 82 | verified | ||
| 11 | Q Qwen3.6-35B-A3B🇨🇳 | 81.7 | Alibaba Cloud / Qwen Team | verified |
| 12 | 81.6 | OpenAI | verified | |
| 13 | Q Qwen3.5-35B-A3B🇨🇳 | 81.4 | Alibaba Cloud / Qwen Team | verified |
| 14 | 79.7 | verified | ||
| 15 | 79.6 | verified | ||
| 16 | S Step3-VL-10B🇨🇳 | 78.1 | StepFun | verified |
| 17 | X Grok-3🇺🇸 | 78 | xAI | verified |
| 18 | 77.6 | OpenAI | verified | |
| 19 | 75.4 | verified | ||
| 20 | 75.2 | OpenAI | verified | |
| 21 | 75.1 | Cohere | verified | |
| 22 | 75 | Anthropic | verified | |
| 23 | 74.8 | OpenAI | verified | |
| 24 | 74.4 | Anthropic | verified | |
| 25 | 74.2 | Anthropic | ||
| 26 | M Llama 4 Maverick🇺🇸 | 73.4 | Meta | verified |
| 27 | 72.9 | verified | ||
| 28 | 72.7 | OpenAI | verified | |
| 29 | 72.5 | Anthropic | ||
| 30 | 72.2 | OpenAI | verified | |
| 31 | 70.7 | verified | ||
| 32 | Q QvQ-72B-Preview🇨🇳 | 70.3 | Alibaba Cloud / Qwen Team | verified |
| 33 | Q Qwen2.5 VL 72B Instruct🇨🇳 | 70.2 | Alibaba Cloud / Qwen Team | verified |
| 34 | Q Qwen2.5 VL 32B Instruct🇨🇳 | 70 | Alibaba Cloud / Qwen Team | verified |
| 35 | M Kimi-k1.5🇨🇳 | 70 | Moonshot AI | verified |
| 36 | M Llama 4 Scout🇺🇸 | 69.4 | Meta | verified |
| 37 | 69.1 | OpenAI | ||
| 38 | 68.3 | Anthropic | verified | |
| 39 | 68.3 | Anthropic | ||
| 40 | 68 | verified | ||
| 41 | X Grok-2🇺🇸 | 66.1 | xAI | verified |
| 42 | 65.9 | verified | ||
| 43 | 64 | Mistral AI | verified | |
| 44 | X Grok-2 mini🇺🇸 | 63.2 | xAI | verified |
| 45 | 62.5 | Mistral AI | verified | |
| 46 | 62.3 | verified | ||
| 47 |  Nova Pro🇺🇸 | 61.7 | Amazon | verified |
| 48 | M Llama 3.2 90B Instruct🇺🇸 | 60.3 | Meta | verified |
| 49 | 59.4 | OpenAI | verified | |
| 50 | 59.3 | Mistral AI | verified | |
| 51 | 59.3 | Mistral AI | verified | |
| 52 | Q Qwen2.5-Omni-7B🇨🇳 | 59.2 | Alibaba Cloud / Qwen Team | verified |
| 53 | Q Qwen2.5 VL 7B Instruct🇨🇳 | 58.6 | Alibaba Cloud / Qwen Team | verified |
| 54 | Nova Lite🇺🇸 | 56.2 | Amazon | verified |
| 55 | 55.4 | OpenAI | verified | |
| 56 | 55.1 | Microsoft | verified | |
| 57 | 53.7 | verified | ||
| 58 | X Grok-1.5V🇺🇸 | 53.6 | xAI | verified |
| 59 | X Grok-1.5🇺🇸 | 53.6 | xAI | verified |
| 60 | 52.5 | Mistral AI | verified | |
| 61 | 51.1 | DeepSeek | verified | |
| 62 | M Llama 3.2 11B Instruct🇺🇸 | 50.7 | Meta | verified |
| 63 | 48 | DeepSeek | verified | |
| 64 | 47.9 | verified | ||
| 65 | 43 | Microsoft | verified | |
| 66 | 40.7 | DeepSeek | verified | |
| 67 | 0 | OpenAI | verified |