WebApp1K: leaderboard
React web app code generation benchmark: 1,000 single-task problems across 20 scenarios (blogging, ecommerce, weather, etc.) testing functional correctness with pass@1.
Metric: Pass@1 (%). Source: arxiv.org. Source: huggingface.co. Saturation forecast: Estimated already saturated. 34 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 Mini | 96.1 | #272 |
| 2 | O1 Preview | 95.2 | #193 |
| 3 | O1 Mini | 93.9 | #323 |
| 4 | DeepSeek R1 | 92.7 | #251 |
| 5 | GPT-4o (2024-08-06) | 88.5 | #336 |
| 6 | Claude 3.5 Sonnet (20240620) | 88.08 | #299 |
| 7 | DeepSeek V3 | 87.23 | #321 |
| 8 | GPT-4o (2024-05-13) | 87.02 | #325 |
| 9 | QwQ-32B | 87 | #421 |
| 10 | Gemini 2.0 Flash (Thinking) | 85.9 | #342 (Gemini 2.0 Flash) |
| 11 | DeepSeek V2.5 | 83.38 | #489 |
| 12 | GPT-4o Mini | 82.71 | #599 |
| 13 | Gemini 2.0 Flash | 82.2 | #342 |
| 14 | Mistral Large 2 (Jul) | 78.04 | #484 |
| 15 | Llama 4 Maverick | 76.9 | #464 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=webapp1k · How It Works · Data refreshed daily, snapshot 2026-10-20.