Cultural Understanding Evaluation on VULCA-BENCH
4.02S*_II ScoreGemini-2.5-Pro
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Gemini-2.5-Pro2026.01 | 4.02 | 4.5 | 4.47 | 4.5 | 4.45 | 4.48 | 3.69 | |
| Qwen3-VL-235B2026.01 | 3.99 | 4.5 | 4.45 | 4.5 | 4.43 | 4.48 | 3.71 | |
| Claude-Sonnet-4.52026.01 | 3.87 | 4.5 | 4.48 | 4.5 | 4.48 | 4.5 | 3.61 | |
| Gemini-3-Pro2026.01 | 3.78 | 4.5 | 4.44 | 4.51 | 4.48 | 4.49 | 3.37 | |
| Llama4-Scout2026.01 | 3.7 | 4.5 | 4.36 | 4.48 | 4.37 | 4.46 | 3.83 | |
| GPT-52026.01 | 3.65 | 4.44 | 4.39 | 4.42 | 4.4 | 4.43 | 3.18 | |
| Claude-Opus-4.52026.01 | 3.65 | 4.5 | 4.44 | 4.5 | 4.46 | 4.49 | 3.24 | |
| GPT-5.22026.01 | 3.64 | 4.5 | 4.4 | 4.51 | 4.41 | 4.5 | 3.33 | |
| Qwen3-VL-Flash2026.01 | 3.53 | 4.5 | 4.54 | 4.48 | 4.56 | 4.48 | 3.51 | |
| GPT-5-mini2026.01 | 3.5 | 4.5 | 4.38 | 4.5 | 4.43 | 4.49 | 3.21 | |
| Claude-Haiku-3.52026.01 | 3.49 | 4.5 | 4.47 | 4.47 | 4.45 | 4.47 | 3.21 | |
| GPT-4o2026.01 | 3.47 | 4.5 | 4.42 | 4.48 | 4.37 | 4.45 | 3.22 | |
| GPT-4o-mini2026.01 | 3.35 | 4.5 | 4.33 | 4.44 | 4.36 | 4.44 | 3.2 | |
| GLM-4V-Flash2026.01 | 3.31 | 4.49 | 4.22 | 4.35 | 4.29 | 4.34 | 3.5 | |
| DeepSeek-VL22026.01 | 3.24 | 4.42 | 4.02 | 4.15 | 4.05 | 4.17 | 3.48 |