Individual benchmark scores plotted by date.
| Organisation | Model | Reported | Top Score | Info | Self Reported | Source |
|---|---|---|---|---|---|---|
| Kimi K2.5 | 27 Jan 2026 | 92.60 | - | Yes | Source | |
| Qwen 3 VL 235B A22B Thinking | - | 89.50 | - | Yes | - | |
| Qwen 3 VL 235B A22B Instruct | - | 89.20 | - | Yes | - | |
| Qwen 3 VL 32B Thinking | - | 89.20 | - | Yes | - | |
| Qwen 3 VL 32B Instruct | - | 87.00 | - | Yes | - | |
| Qwen 3 VL 8B Thinking | - | 86.00 | - | Yes | - | |
| Qwen 3 8B | - | 86.00 | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50) | Yes | - | |
| Qwen 3 Embedding 8B | - | 86.00 | inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50) | Yes | - | |
| Qwen 3 Guard Gen 8B | - | 86.00 | inferred family alias from qwen3-vl-8b-thinking (score=0.3400; benches=50) | Yes | - | |
| Qwen 3 Guard Stream 8B | - | 86.00 | inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50) | Yes | - | |
| Qwen 3 Reranker 8B | - | 86.00 | inferred family alias from qwen3-vl-8b-thinking (score=0.3850; benches=50) | Yes | - | |
| Qwen 3 VL 30B A3B Thinking | - | 86.00 | - | Yes | - | |
| Qwen 3 VL Embedding 8B | - | 86.00 | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50) | Yes | - | |
| Qwen 3 VL Reranker 8B | - | 86.00 | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50) | Yes | - | |
| Qwen 2 VL 72B | - | 84.50 | - | Yes | Source | |
| Qwen 2 VL 2B | - | 84.50 | inferred family alias from qwen2-vl-72b (score=0.4083; benches=15) | Yes | Source | |
| Qwen 3 VL 8B Instruct | - | 83.10 | - | Yes | - | |
| Qwen 3 VL 4B Thinking | - | 83.00 | - | Yes | - | |
| Qwen 3 4B Instruct 2507 | - | 83.00 | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Yes | - | |
| Qwen 3 4B SafeRL | - | 83.00 | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 4B Thinking 2507 | - | 83.00 | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Yes | - | |
| Qwen 3 Embedding 4B | - | 83.00 | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 Guard Gen 4B | - | 83.00 | inferred family alias from qwen3-vl-4b-thinking (score=0.3400; benches=48) | Yes | - | |
| Qwen 3 Guard Stream 4B | - | 83.00 | inferred family alias from qwen3-vl-4b-thinking (score=0.3371; benches=48) | Yes | - | |
| Qwen 3 Reranker 4B | - | 83.00 | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Yes | - | |
| Qwen 3 VL 30B A3B Instruct | - | 82.00 | - | Yes | - | |
| Qwen 3 VL 4B Instruct | - | 80.30 | - | Yes | - |