I-WebGenBench - Computer Science: leaderboard
I-WebGenBench evaluates whether a model can generate an interactive scientific web application from a paper-derived specification across five domains (biology, chemistry, computer science, economics, human-computer interaction), with 201 tasks. This board reports Overall rubric score (%) on the computer-science domain tasks; higher is better.
Metric: Overall rubric score (%) on the computer-science domain tasks; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 76.1 | #123 |
| 2 | Kimi K2.5 | 51.8 | #256 |
| 3 | Gemini 3.1 Pro (Preview) | 45.2 | #94 |
| 4 | MiniMax-M2.7 | 35.1 | #518 |
| 5 | Qwen 3 Max | 32.5 | #403 |
| 6 | DeepSeek V4 Flash | 21.6 | #258 |
| 7 | MiMo-V2.5 | 17.2 | #354 |
| 8 | DeepSeek V3.2 | 8.8 | #406 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=i-webgenbench-computer-science · How It Works · Data refreshed daily, snapshot 2026-11-13.