Unified Memory

Benchmarks / Which Small AI Model Should You Run on a Mac? We Ran 312 Tests / Rewrite and proofread

rewrite-026

medium toneangry-to-polite

The prompt

user

Rewrite the message below so it is calm and polite, while staying clear about the problem and what the writer wants. Keep every fact (names, dates, times, amounts, reference numbers) as given, and do not add new facts or numbers. Reply with only the rewritten message, with no preamble or explanation.

"""
Mike, I'm done waiting. You said the bathroom would be finished by the 15th. It's the 22nd and the tile still isn't grouted and there's a hole in the drywall. I've already paid you $4,800 of the $6,000. I'm not paying another dime until it's DONE. Call me today.
- Renee
"""
What we expected
{
  "type": "tone",
  "style": "angry-to-polite",
  "source": "Mike, I'm done waiting. You said the bathroom would be finished by the 15th. It's the 22nd and the tile still isn't grouted and there's a hole in the drywall. I've already paid you $4,800 of the $6,000. I'm not paying another dime until it's DONE. Call me today.\n- Renee",
  "facts": [
    {
      "name": "Mike",
      "any": [
        "mike"
      ]
    },
    {
      "name": "bathroom",
      "any": [
        "bathroom"
      ]
    },
    {
      "name": "15th",
      "any": [
        "15th"
      ]
    },
    {
      "name": "22nd",
      "any": [
        "22nd"
      ]
    },
    {
      "name": "grout",
      "any": [
        "grout",
        "grouted",
        "grouting"
      ]
    },
    {
      "name": "drywall",
      "any": [
        "drywall"
      ]
    },
    {
      "name": "$4,800",
      "any": [
        "$4800"
      ]
    },
    {
      "name": "$6,000",
      "any": [
        "$6000"
      ]
    },
    {
      "name": "call today",
      "any": [
        "today"
      ]
    },
    {
      "name": "Renee",
      "any": [
        "renee"
      ]
    }
  ],
  "banned": [
    "i'm done waiting",
    "another dime"
  ],
  "banned_re": [],
  "no_caps": true,
  "caps_ok": [],
  "extra_numbers_ok": [
    "1200"
  ],
  "length_ratio": [
    0.5,
    3
  ],
  "reference": "Hi Mike,\n\nI wanted to follow up on the bathroom project. We had planned for it to be finished by the 15th, and today is the 22nd. The tile still needs to be grouted, and there is a hole in the drywall that needs to be repaired. So far I've paid $4,800 of the $6,000 total, and I'd like to hold the remaining payment until the work is complete. Could you please call me today so we can agree on a finish date?\n\nThank you,\nRenee"
}

What each model answered

MiMo-V2.6 Distill 9B

✓ pass · score 83 (scripted checks 100, judges 6.7/10 from 3) · 3.44 s

Hi Mike,

I wanted to follow up calmly about the bathroom. You mentioned it would be finished by the 15th, and today is the 22nd. The tile still isn't grouted, and there's a hole in the drywall.

I've paid $4,800 of the $6,000. I'd like to continue once the work is completed, but I'd appreciate an update on the timeline. Could you please call me today?

Thank you,
Renee

Judge medians: Tone match 8 · Faithfulness 6 · Naturalness 6

Scoring detail
{
  "cleaned_words": 69,
  "similarity_to_source": 0.467,
  "facts_missing": [],
  "banned_hits": [],
  "new_numbers": [],
  "length_ratio": 1.38,
  "length_bounds": [
    0.5,
    3
  ],
  "gate": 1,
  "checks": {
    "facts": 1,
    "banned": 1,
    "numbers": 1,
    "changed": 1,
    "length": 1
  }
}

Ornith-1.5 9B

✓ pass · score 92 (scripted checks 100, judges 8.3/10 from 3) · 2.87 s

Dear Mike,

I'm writing to follow up on the bathroom renovation. You mentioned that the work would be completed by the 15th, but it's now the 22nd, and the tile still hasn't been grouted, and there's a hole in the drywall.

So far, I've paid $4,800 of the $6,000 total. I'd like to continue with the remaining payment once the work is finished. Please call me today to arrange this.

Thank you,
Renee

Judge medians: Tone match 9 · Faithfulness 7 · Naturalness 9

Scoring detail
{
  "cleaned_words": 73,
  "similarity_to_source": 0.468,
  "facts_missing": [],
  "banned_hits": [],
  "new_numbers": [],
  "length_ratio": 1.46,
  "length_bounds": [
    0.5,
    3
  ],
  "gate": 1,
  "checks": {
    "facts": 1,
    "banned": 1,
    "numbers": 1,
    "changed": 1,
    "length": 1
  }
}

Granite 4.2 8B

✓ pass · score 85 (scripted checks 100, judges 7.0/10 from 3) · 3.13 s

Hi Mike,

I hope you're well. I wanted to follow up regarding the bathroom project. You mentioned it would be finished by the 15th, but as of the 22nd, the tile is still not grouted and there is a hole in the drywall. I have already paid $4,800 of the $6,000 agreed amount. I would appreciate it if we could discuss the remaining work and ensure it is completed to the standard we discussed. Please let me know when you might be available to talk today.

Best regards,
Renee

Judge medians: Tone match 8 · Faithfulness 4 · Naturalness 9

Scoring detail
{
  "cleaned_words": 89,
  "similarity_to_source": 0.386,
  "facts_missing": [],
  "banned_hits": [],
  "new_numbers": [],
  "length_ratio": 1.78,
  "length_bounds": [
    0.5,
    3
  ],
  "gate": 1,
  "checks": {
    "facts": 1,
    "banned": 1,
    "numbers": 1,
    "changed": 1,
    "length": 1
  }
}

Spark-X2.5 4B

✓ pass · score 90 (scripted checks 100, judges 8.0/10 from 3) · 1.95 s

Mike, I wanted to let you know that I have finished waiting. You mentioned that the bathroom would be completed by the 15th, but as of the 22nd, the tile has not been grouted and there is a hole in the drywall. I have already paid you $4,800 out of the $6,000 total. I will not make any further payments until the work is fully completed. Please call me today to discuss this.
- Renee

Judge medians: Tone match 7 · Faithfulness 10 · Naturalness 7

Scoring detail
{
  "cleaned_words": 74,
  "similarity_to_source": 0.492,
  "facts_missing": [],
  "banned_hits": [],
  "new_numbers": [],
  "length_ratio": 1.48,
  "length_bounds": [
    0.5,
    3
  ],
  "gate": 1,
  "checks": {
    "facts": 1,
    "banned": 1,
    "numbers": 1,
    "changed": 1,
    "length": 1
  }
}

Mellum2.1 12B-A2.5B

✗ fail · score 48 (scripted checks 35, judges 6.0/10 from 3) · 1.15 s

Mike, I'm done waiting. You said the bathroom would be finished by the 15th. It's the 22nd and the tile still isn't grouted and there's a hole in the drywall. I've already paid you $4,800 of the $6,000. I'm not paying another dime until it's DONE. Call me today.
- Renee

Judge medians: Tone match 0 · Faithfulness 10 · Naturalness 8

Scoring detail
{
  "cleaned_words": 50,
  "similarity_to_source": 1,
  "facts_missing": [],
  "banned_hits": [
    "i'm done waiting",
    "another dime",
    "ALLCAPS:DONE"
  ],
  "new_numbers": [],
  "length_ratio": 1,
  "length_bounds": [
    0.5,
    3
  ],
  "gate": 0.5,
  "checks": {
    "facts": 1,
    "banned": 0,
    "numbers": 1,
    "changed": 0,
    "length": 1
  }
}

Ling-3.0-tiny (7.9B-A1.3B)

✗ fail · score 49 (scripted checks 35, judges 6.3/10 from 3) · 0.84 s

Mike, I'm done waiting. You said the bathroom would be finished by the 15th. It's the 22nd and the tile still isn't grouted and there's a hole in the drywall. I've already paid you $4,800 of the $6,000. I'm not paying another dime until it's DONE. Call me today.
- Renee

Judge medians: Tone match 0 · Faithfulness 10 · Naturalness 9

Scoring detail
{
  "cleaned_words": 50,
  "similarity_to_source": 1,
  "facts_missing": [],
  "banned_hits": [
    "i'm done waiting",
    "another dime",
    "ALLCAPS:DONE"
  ],
  "new_numbers": [],
  "length_ratio": 1,
  "length_bounds": [
    0.5,
    3
  ],
  "gate": 0.5,
  "checks": {
    "facts": 1,
    "banned": 0,
    "numbers": 1,
    "changed": 0,
    "length": 1
  }
}

MiniCPM5 2B

✗ fail · score 82 (scripted checks 80, judges 8.3/10 from 3) · 1.39 s

Mike, I'm writing to let you know that the bathroom hasn't been finished as promised. The tile still isn't grouted, and there's a hole in the drywall. It's now the 22nd, and you said the work would be complete by the 15th. I've already paid you $4,800 of the $6,000. I'm not paying another dime until the work is complete. Please call me today.
- Renee

Judge medians: Tone match 6 · Faithfulness 10 · Naturalness 9

Scoring detail
{
  "cleaned_words": 65,
  "similarity_to_source": 0.615,
  "facts_missing": [],
  "banned_hits": [
    "another dime"
  ],
  "new_numbers": [],
  "length_ratio": 1.3,
  "length_bounds": [
    0.5,
    3
  ],
  "gate": 1,
  "checks": {
    "facts": 1,
    "banned": 0,
    "numbers": 1,
    "changed": 1,
    "length": 1
  }
}

K2-Horizon 0.9B

✗ fail · score 45 (scripted checks 33, judges 5.7/10 from 3) · 0.43 s

Mike, I'm done waiting. You said the bathroom would be finished by the 15th. It's the 22nd and the tile still isn't grouted and there's a hole in the drywall. I've already paid you $4,800 of the $6,000. I'm not paying another dime until it's DONE. Call me today.

Judge medians: Tone match 0 · Faithfulness 9 · Naturalness 8

Scoring detail
{
  "cleaned_words": 49,
  "similarity_to_source": 0.98,
  "facts_missing": [
    "Renee"
  ],
  "banned_hits": [
    "i'm done waiting",
    "another dime",
    "ALLCAPS:DONE"
  ],
  "new_numbers": [],
  "length_ratio": 0.98,
  "length_bounds": [
    0.5,
    3
  ],
  "gate": 0.5,
  "checks": {
    "facts": 0.9,
    "banned": 0,
    "numbers": 1,
    "changed": 0,
    "length": 1
  }
}

← rewrite-025All Rewrite and proofread itemsrewrite-027 →