MMMT Bench
AI 基准测试结果和模型表现
类型: percentage
已记录的结果
34
平均分数
17.50%
分数范围
6% - 84.90%
领先模型
84.90% - Ministral 3.0 14B
分数变化趋势
按日期绘制的单项基准测试分数。
模型结果
| 组织 | 模型 | 报告时间 | 最高分数 | 信息 | 自行报告 | 来源 |
|---|---|---|---|---|---|---|
| Ministral 3.0 14B | 2025年12月02日 | 84.90% | Multimodal | 是 | 来源 | |
| Ministral 3.0 8B | 2025年12月02日 | 80.80% | Multimodal | 是 | 来源 | |
| Ministral 3.0 3B | 2025年12月02日 | 78.30% | Multimodal | 是 | 来源 | |
| Pixtral Large | 2024年11月18日 | 74% | - | 是 | 来源 | |
| Pixtral 12B | 2024年9月17日 | 60.50% | inferred version-family alias from pixtral-12b-2409 | 是 | 来源 | |
| Qwen 3 VL 235B A22B Instruct | - | 8.50% | - | 是 | - | |
| Qwen 3 VL 235B A22B Thinking | - | 8.50% | - | 是 | - | |
| Qwen 3 VL 32B Instruct | - | 8.40% | - | 是 | - | |
| Qwen 3 VL 32B Thinking | - | 8.30% | - | 是 | - | |
| Qwen 3 VL 30B A3B Instruct | - | 8.10% | - | 是 | - | |
| Qwen 3 8B | - | 8% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50) | 是 | - | |
| Qwen 3 Guard Stream 8B | - | 8% | inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50) | 是 | - | |
| Qwen 3 VL 8B Thinking | - | 8% | - | 是 | - | |
| Qwen 3 VL Embedding 8B | - | 8% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50) | 是 | - | |
| Qwen 3 VL Reranker 8B | - | 8% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50) | 是 | - | |
| Qwen 3 VL 30B A3B Thinking | - | 7.90% | - | 是 | - | |
| Qwen 3 4B Instruct 2507 | - | 7.70% | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | 是 | - | |
| Qwen 3 4B SafeRL | - | 7.70% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | 是 | - | |
| Qwen 3 4B Thinking 2507 | - | 7.70% | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | 是 | - | |
| Qwen 3 Embedding 4B | - | 7.70% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | 是 | - | |
| Qwen 3 Guard Gen 4B | - | 7.70% | inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48) | 是 | - | |
| Qwen 3 Guard Stream 4B | - | 7.70% | inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48) | 是 | - | |
| Qwen 3 Reranker 4B | - | 7.70% | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | 是 | - | |
| Qwen 3 VL 4B Thinking | - | 7.70% | - | 是 | - | |
| Qwen 3 VL 8B Instruct | - | 7.70% | - | 是 | - |