Multimodal Data Synthesis on DataClaw0 GUI (val)
100Field ScoreGPT-4o (1120-global)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-4o (1120-global)2026.06 | 100 | 84.81 | 80.69 | |
| Gemini-1.5-Pro-Preview2026.06 | 100 | 90.01 | 99.67 | |
| DataClaw0-O2026.06 | 100 | 80.01 | 85.7 | |
| DataClaw0-E2026.06 | 100 | 89.18 | 96.33 | |
| Qwen2.5-7B2026.06 | 94.87 | 72.72 | 72.71 | |
| MiniMax-v2.72026.06 | 92.5 | 78.93 | 77.86 | |
| Qwen2.5-plus2026.06 | 70 | 62.64 | 66.96 | |
| Claude-Sonnet-3.52026.06 | 57.5 | 50.05 | 54.06 |