AI 基准测试结果和模型表现
按日期绘制的单项基准测试分数。
| 组织 | 模型 | 报告时间 | 最高分数 | 信息 | 自行报告 | 来源 |
|---|---|---|---|---|---|---|
| Qwen 3.6 Plus | 2026年4月01日 | 83.40% | - | 是 | 来源 | |
| Qwen 3 VL 235B A22B Instruct | - | 82.20% | - | 是 | - | |
| Qwen 3.5 122B A10B | 2026年2月24日 | 81.80% | - | 是 | 来源 | |
| Qwen 3 VL 235B A22B Thinking | - | 81.50% | - | 是 | - | |
| Qwen 3.5 27B | 2026年2月24日 | 81% | - | 是 | 来源 | |
| Qwen 3.5 Flash | 2026年2月23日 | 81% | inferred family alias from qwen3.5-27b (score=0.4147; benches=81) | 是 | 来源 | |
| Qwen 3 VL 30B A3B Instruct | - | 80.70% | - | 是 | - | |
| Qwen 3.5 35B A3B | 2026年2月24日 | 80.70% | - | 是 | 来源 | |
| Qwen 3 VL 32B Instruct | - | 80.30% | - | 是 | - | |
| Qwen 3 VL 8B Instruct | - | 79.90% | - | 是 | - | |
| Qwen 2.5 Math 72B | - | 79.80% | inferred high-confidence family alias from qwen2.5-vl-72b (score=0.4700; benches=30) | 是 | 来源 | |
| Qwen 2.5 Math PRM 72B | - | 79.80% | inferred family alias from qwen2.5-vl-72b (score=0.4050; benches=30) | 是 | 来源 | |
| Qwen 2.5 Math RM 72B | - | 79.80% | inferred family alias from qwen2.5-vl-72b (score=0.4092; benches=30) | 是 | 来源 | |
| Qwen 2.5 VL 72B Instruct | 2025年1月26日 | 79.80% | - | 是 | 来源 | |
| Qwen 2.5 0.5B | - | 77.80% | inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32) | 是 | 来源 | |
| Qwen 2.5 1.5B | - | 77.80% | inferred family alias from qwen2.5-vl-7b (score=0.3131; benches=32) | 是 | 来源 | |
| Qwen 2.5 3B | - | 77.80% | inferred family alias from qwen2.5-vl-7b (score=0.3531; benches=32) | 是 | 来源 | |
| Qwen 2.5 VL 7B Instruct | - | 77.80% | - | 是 | 来源 | |
| Qwen 3 VL 30B A3B Thinking | - | 77.80% | - | 是 | - | |
| Qwen 2.5 Coder 32B Instruct | - | 77.10% | inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4641; benches=28) | 是 | 来源 | |
| Qwen 2.5 VL 32B Instruct | 2025年3月24日 | 77.10% | - | 是 | 来源 | |
| Qwen 2.5 VL 3B Instruct | - | 77.10% | inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4914; benches=28) | 是 | 来源 | |
| Qwen 3 8B | - | 76.30% | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50) | 是 | - | |
| Qwen 3 Guard Stream 8B | - | 76.30% | inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50) | 是 | - | |
| Qwen 3 VL 8B Thinking | - | 76.30% | - | 是 | - |