Website Generation on WebGen-Bench
57.72FSRClaude-3.7-Sonnet
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Claude-3.7-SonnetInstitution=Anthropic2026.04 | 57.72 | 3.9 | 84 | |
| GPT-5Institution=OpenAI2026.04 | 46.53 | 3.34 | 90.43 | |
| Claude-Sonnet-4Institution=Anthropic2026.04 | 46.13 | 3.86 | 86.05 | |
| GPT-4.1Institution=OpenAI2026.04 | 43.91 | 3.78 | 82.09 | |
| o3Institution=OpenAI2026.04 | 42.86 | 3.55 | 81.08 | |
| Gemini-2.5-ProInstitution=Google2026.04 | 36.31 | 3.89 | 83.33 | |
| DeepSeek-R1Institution=DeepSeek2026.04 | 30.25 | 3.67 | 42.86 | |
| WebGen-R1-7BInstitution=Ours2026.04 | 29.21 | 3.94 | 95.89 | |
| o4-miniInstitution=OpenAI2026.04 | 27.29 | 3.31 | 56.52 | |
| GPT-4oInstitution=OpenAI2026.04 | 21.6 | 3.31 | 85.71 | |
| Qwen3-32BInstitution=Alibaba2026.04 | 18.69 | 3.39 | 59.42 | |
| Qwen3-30B-A3B-Thinking-2507Institution=Alibaba2026.04 | 9.3 | 2.6 | 23.81 | |
| Qwen3-Coder-30B-A3B-InstructInstitution=Alibaba2026.04 | 6.06 | 2.9 | 32.81 | |
| Qwen3-8BInstitution=Alibaba2026.04 | 3.72 | 2.57 | 12.5 | |
| Qwen2.5-72B-InstructInstitution=Alibaba2026.04 | 2.54 | 3.14 | 8.86 | |
| Qwen2.5-Coder-7B-InstructInstitution=Alibaba2026.04 | 1.59 | 2.73 | 30.56 |