Singapore knowledge evaluation on SG-Legal
86.4Pass@1Phoenix-VL 1.5 Medium
Evaluation Results
| Method | Links | |
|---|---|---|
| Phoenix-VL 1.5 MediumZero-shot=true, Parameter count=123B, Organization=HTX2026.05 | 86.4 | |
| Qwen 3.5Zero-shot=true, Parameter count=122B-A10B, Organization=Alibaba Cloud2026.05 | 61.44 | |
| Nemotron 3 SuperZero-shot=true, Parameter count=120B-A12B, Organization=Nvidia2026.05 | 59.19 | |
| GPT-OSSZero-shot=true, Parameter count=117B-A5B, Organization=OpenAI2026.05 | 55.23 | |
| Llama 4 MaverickZero-shot=true, Parameter count=400B-A17B, Organization=MetaAI2026.05 | 50.54 | |
| GLM-4.5VZero-shot=true, Parameter count=106B-A12B, Organization=Z.ai2026.05 | 48.83 |