GUI Generation on Android Control (in-domain)
94.28SadCode2World-8B
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Code2World-8BModel Category=Code Generation Models2026.02 | 94.28 | 88.64 | 71.35 | 70.32 | 79.44 | 49.18 | |
| GPT-5Model Category=Code Generation Models2026.02 | 94.02 | 90.15 | 74.13 | 69.78 | 78.07 | 41.74 | |
| Gemini-3-FlashModel Category=Code Generation Models2026.02 | 92.65 | 84.08 | 74.52 | 69.74 | 81.17 | 45.72 | |
| Gemini-3-Pro-ImageModel Category=Image Generation Models2026.02 | 92.63 | 83.67 | 68.47 | 63.67 | 84.89 | 64.11 | |
| GLM-4.6V-106BModel Category=Code Generation Models2026.02 | 91.62 | 74.23 | 61.95 | 58.74 | 67.26 | 27.42 | |
| Claude-Sonnet-4.5Model Category=Code Generation Models2026.02 | 89.6 | 86.12 | 65.8 | 62.25 | 75.97 | 36.72 | |
| GPT-Image-1Model Category=Image Generation Models2026.02 | 89.59 | 71.4 | 58.78 | 56.22 | 77.58 | 61 | |
| Doubao-Seedream-4.5Model Category=Image Generation Models2026.02 | 85.36 | 86.15 | 59.08 | 55.82 | 81.76 | 52.19 | |
| Qwen2.5-VL-72BModel Category=Code Generation Models2026.02 | 73.34 | 70.12 | 47.15 | 48.26 | 68.45 | 30.12 | |
| InternVL3-78BModel Category=Code Generation Models2026.02 | 72.41 | 67.35 | 45.06 | 46.73 | 62.64 | 25.42 | |
| Qwen3-VL-8BModel Category=Code Generation Models2026.02 | 59.2 | 65.8 | 43.1 | 42.7 | 63.88 | 28.86 | |
| Janus-Pro-7BModel Category=Image Generation Models2026.02 | 58.1 | 53.68 | 55.22 | 45.9 | 79.85 | 54.3 | |
| Qwen-Image-Edit-MaxModel Category=Image Generation Models2026.02 | 57.55 | 54.12 | 54.05 | 46.33 | 70.39 | 53.61 | |
| JanusCoderV-7BModel Category=Code Generation Models2026.02 | 57.12 | 56.05 | 30.18 | 31.09 | 59.89 | 21.73 |