Structure & Data · blind round

R147 · CSV with embedded comma to JSON array

blind round 11/11 local models answered

Every answer below carries the name of the model that wrote it. Blind refers to the scoring: the review was done without knowing which model produced which answer — the published result names them.

Show task

Convert the following CSV into a JSON array of objects, using the header row as field names. Keep "id" and "amount" as numbers; keep "name" and "notes" as strings. Watch out for the quoted field that contains a comma inside it. Output ONLY the JSON array — no explanation.

id,name,notes,amount
1,Anna Berg,"Prefers email, not phone",120.50
2,Tom Klein,Standard customer,75.00
3,Lena Fuchs,"Called twice, resolved",200.00
4,Max Otto,VIP customer,50.00

The answers

Local models (Brain cluster) — identical prompt, shuffled order (the sequence is not a ranking)

qwen3:8b

[{"id":1,"name":"Anna Berg","notes":"Prefers email, not phone","amount":120.5},{"id":2,"name":"Tom Klein","notes":"Standard customer","amount":75},{"id":3,"name":"Lena Fuchs","notes":"Called twice, resolved","amount":200},{"id":4,"name":"Max Otto","notes":"VIP customer","amount":50}]

deepseek-r1:14b

[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.50
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.00
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.00
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.00
  }
]

mistral-small:24b

```json
[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.50
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.00
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.00
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.00
  }
]
```

qwen3-coder:30b

[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.5
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.0
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.0
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.0
  }
]

llama3.1:8b

```json
[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.5
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.0
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.0
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.0
  }
]
```

gemma4:26b

```json
[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.5
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.0
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.0
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.0
  }
]
```

qwen3.5:9b

```json
[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.5
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.0
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.0
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.0
  }
]
```

qwen3:14b

```json
[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.50
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.00
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.00
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.00
  }
]
```

command-r:35b

[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.5
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50
  }
]

deepseek-r1:32b

[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.5
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50
  }
]

qwq:32b

[
  {"id":1,"name":"Anna Berg","notes":"Prefers email, not phone","amount":120.50},
  {"id":2,"name":"Tom Klein","notes":"Standard customer","amount":75.00},
  {"id":3,"name":"Lena Fuchs","notes":"Called twice, resolved","amount":200.00},
  {"id":4,"name":"Max Otto","notes":"VIP customer","amount":50.00}
]

Outside the Brain cluster — same task, same character budget, collected through their own tooling (response times not comparable)

Claude Opus 5frontier reference

[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.50
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.00
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.00
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.00
  }
]

Claude Sonnet 5frontier reference

[
  {"id": 1, "name": "Anna Berg", "notes": "Prefers email, not phone", "amount": 120.50},
  {"id": 2, "name": "Tom Klein", "notes": "Standard customer", "amount": 75.00},
  {"id": 3, "name": "Lena Fuchs", "notes": "Called twice, resolved", "amount": 200.00},
  {"id": 4, "name": "Max Otto", "notes": "VIP customer", "amount": 50.00}
]

Claude Fable 5frontier reference

[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.50
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.00
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.00
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.00
  }
]

Claude Haiku 4.5frontier reference

[
  {
    "id": 1,
    "name": "Anna Berg",
    "notes": "Prefers email, not phone",
    "amount": 120.50
  },
  {
    "id": 2,
    "name": "Tom Klein",
    "notes": "Standard customer",
    "amount": 75.00
  },
  {
    "id": 3,
    "name": "Lena Fuchs",
    "notes": "Called twice, resolved",
    "amount": 200.00
  },
  {
    "id": 4,
    "name": "Max Otto",
    "notes": "VIP customer",
    "amount": 50.00
  }
]

Reference

Gold — the correct answerreference

[{"id":1,"name":"Anna Berg","notes":"Prefers email, not phone","amount":120.5},{"id":2,"name":"Tom Klein","notes":"Standard customer","amount":75.0},{"id":3,"name":"Lena Fuchs","notes":"Called twice, resolved","amount":200.0},{"id":4,"name":"Max Otto","notes":"VIP customer","amount":50.0}] — maschinell verifiziert (Python csv.DictReader + json.loads Round-Trip, 4 Objekte, quoted-Comma-Felder korrekt als je EIN Feld erhalten, nicht aufgesplittet).

Results at a glance— every model in this round with score, response time and status.

Who wrote what

Model	Blind score	Latency	Status
qwen3:8b	10.0	36.4 s	OK
deepseek-r1:14b	10.0	30.4 s	OK
mistral-small:24b	7.0	59.8 s	OK
qwen3-coder:30b	10.0	9.8 s	OK
llama3.1:8b	7.0	46.8 s	OK
gemma4:26b	7.0	104.9 s	OK
qwen3.5:9b	7.0	20.5 s	OK
qwen3:14b	7.0	10.6 s	OK
command-r:35b	10.0	131.0 s	OK
deepseek-r1:32b	10.0	53.8 s	OK
qwq:32b	10.0	117.5 s	OK

Verdict

6 of 11 answering local models matched the gold answer (blind score ≥ 8): qwen3-coder:30b, qwen3:8b, deepseek-r1:14b, deepseek-r1:32b, qwq:32b, command-r:35b.