ScratchWorld - Create (Primitive Mode): leaderboard
Create tasks from ScratchWorld (Zhang et al., 2026): build a working Scratch project from a natural-language specification, checked by unit tests run in the Scratch VM. Primitive mode, where the agent works from screenshots and must assemble the blocks through low-level GUI actions such as click and drag-and-drop. The score is the percentage of the 20 tasks on which every unit test passes; higher is better.
Metric: Success Rate (%). Source: huggingface.co. Saturation forecast: Around October 2027. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 15 | #163 |
| 2 | Claude Sonnet 4.5 | 10 | #244 |
| 3 | Gemini 2.5 Pro | 5 | #262 |
| 4 | Qwen 3 VL 32B Instruct | 0 | #582 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=scratchworld-create-primitive-mode · How It Works · Data refreshed daily, snapshot 2026-11-12.