React UI Generation on Cookie-Bench (Dynamic split)
87.8Functionality ScoreOpus 4.6
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Opus 4.6Generation Mode=React agent scaffold2026.05 | 87.8 | 84.7 | 79.4 | |
| Opus 4.7Generation Mode=HTML direct chat2026.05 | 87.8 | 91.4 | 83.5 | |
| Opus 4.7Generation Mode=React agent scaffold2026.05 | 85.8 | 94.1 | 82.9 | |
| Kimi-K2.6Generation Mode=HTML direct chat2026.05 | 85 | 88.6 | 80.8 | |
| Qwen3.6-PlusGeneration Mode=React agent scaffold2026.05 | 83.9 | 79.9 | 74.8 | |
| Opus 4.6Generation Mode=HTML direct chat2026.05 | 82.2 | 93.2 | 79 | |
| Mimo-V2-ProGeneration Mode=React agent scaffold2026.05 | 82.1 | 84.7 | 75.2 | |
| Kimi-K2.6Generation Mode=React agent scaffold2026.05 | 81.6 | 91.1 | 78.9 | |
| DS-V4-ProGeneration Mode=HTML direct chat2026.05 | 80.9 | 90.4 | 76.8 | |
| GPT-5.4Generation Mode=HTML direct chat2026.05 | 80.9 | 90 | 76.9 | |
| GLM-5.1Generation Mode=HTML direct chat2026.05 | 80.6 | 90.5 | 76.5 | |
| GPT-5.4Generation Mode=React agent scaffold2026.05 | 80.5 | 90.5 | 76.4 | |
| Gemini-3.1-ProGeneration Mode=HTML direct chat2026.05 | 80.5 | 85.7 | 73.7 | |
| Mimo-V2-ProGeneration Mode=HTML direct chat2026.05 | 79.9 | 87.5 | 73.8 | |
| GLM-5.1Generation Mode=React agent scaffold2026.05 | 79.5 | 85.7 | 74.6 | |
| Qwen3.6-PlusGeneration Mode=HTML direct chat2026.05 | 78.7 | 90.3 | 74.3 | |
| DS-V4-FlashGeneration Mode=HTML direct chat2026.05 | 78.1 | 87.1 | 72.4 | |
| DS-V4-ProGeneration Mode=React agent scaffold2026.05 | 76.8 | 87.7 | 72.1 | |
| Qwen3.6-27BGeneration Mode=React agent scaffold2026.05 | 75.6 | 74.2 | 66.3 | |
| Gemini-3-FlashGeneration Mode=HTML direct chat2026.05 | 74.3 | 78.8 | 65.5 | |
| Qwen3.6-27BGeneration Mode=HTML direct chat2026.05 | 73.3 | 87.2 | 69.6 | |
| DS-V4-FlashGeneration Mode=React agent scaffold2026.05 | 71.1 | 79.5 | 66.2 | |
| Gemini-3.1-ProGeneration Mode=React agent scaffold2026.05 | 70.8 | 80.3 | 64.3 | |
| Gemini-3-FlashGeneration Mode=React agent scaffold2026.05 | 68.8 | 82.8 | 63.5 | |
| Qwen3.6-35B-A3BGeneration Mode=HTML direct chat2026.05 | 67.7 | 80.8 | 61.7 | |
| Qwen3.6-35B-A3BGeneration Mode=React agent scaffold2026.05 | 67.3 | 70.1 | 59.8 |