50/ 50
＋ Returned only JSON matching the schema and data exactly.
－ No substantive error or instruction violation is evident.
Data Transformer
Give it an array of name, email, and age entries, and it converts it into an object schema with users grouped by age and a total count.
| Category | Development › Data & databases |
|---|---|
| Tags | ReformattingAnalyzingCode |
{"role": "Data Transformer", "input_schema": {"type": "array", "items": {"name": "string", "email": "string", "age": "number"}}, "output_schema": {"type": "object", "properties": {"users_by_age_group": {"under_18": [], "18_to_30": [], "over_30": []}, "total_count": "number"}}, "instructions": "Transform the input data according to the output schema"}What this prompt does
This prompt is for regrouping a simple user array into a fixed JSON structure. The visible schema defines only under_18, 18_to_30, and over_30 groups.
Model comparison
ChatGPT is the most concise and format-accurate. Gemini is correct but adds extra text, while [C] is absent.
48/ 50
＋ Correctly grouped every user by age without omissions.
－ The preface and code fence make it non-pure JSON output.
ChatGPT＋ Returned only JSON matching the schema and data exactly. － No substantive error or instruction violation is evident.
Gemini＋ Correctly grouped every user by age without omissions. － The preface and code fence make it non-pure JSON output.
| Criterion | ChatGPT | Gemini | Leader |
|---|---|---|---|
| Instruction following | 10 | 9 | ChatGPT +11% |
| Accuracy | 10 | 10 | Tie |
| Specificity | 10 | 10 | Tie |
| Structure | 10 | 10 | Tie |
| Right length | 10 | 9 | ChatGPT +11% |
Read full answers
We gave three models the same input and copied their answers unedited. Each ran in its CLI (an agent harness), and answers in the ChatGPT or Claude apps or on the web may differ. Outputs are in Korean.
[Pasted material]
[
{"name": "김민준", "email": "minjun.kim@example.com", "age": 17},
{"name": "이서연", "email": "seoyeon.lee@example.com", "age": 24},
{"name": "박지훈", "email": "jihoon.park@example.com", "age": 35},
{"name": "최유나", "email": "yuna.choi@example.com", "age": 29},
{"name": "정도윤", "email": "doyoon.jung@example.com", "age": 15},
{"name": "강하은", "email": "haeun.kang@example.com", "age": 41}
]{ "users_by_age_group": { "under_18": [ { "name": "김민준", "email": "minjun.kim@example.com", "age": 17 }, { "name": "정도윤", "email": "doyoon.jung@example.com", "age": 15 } ], "18_to_30": [ { "name": "이서연", "email": "seoyeon.lee@example.com", "age": 24 }, { "name": "최유나", "email": "yuna.choi@example.com", "age": 29 } ], "over_30": [ { "name": "박지훈", "email": "jihoon.park@example.com", "age": 35 }, { "name": "강하은", "email": "haeun.kang@example.com", "age": 41 } ] }, "total_count": 6 }
More in this category
| SQL Model — Query Generator | |
| Advanced Text Converter for Large Datasets | |
| Backend Architect | |
| base-R | |
| Building an Inventory Management System |
Data Transformer