Unified MemoryBenchmarks / Which Small AI Model Should You Run on a Mac? We Ran 312 Tests / Rewrite and proofread
rewrite-024
medium tonecasual-to-professional
The prompt
user
Rewrite the message below in a clear, professional tone suitable for a work email. 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. """ ok so quick update on the warehouse thing - we're kinda behind. the new shelving got here late (like 10 days late ugh) so we're looking at go-live on the 24th instead of the 14th. cost is basically the same, maybe +$3k for overtime. Priya's team is crushing it tho. will ping u friday w/ more """
What we expected
{
"type": "tone",
"style": "casual-to-professional",
"source": "ok so quick update on the warehouse thing - we're kinda behind. the new shelving got here late (like 10 days late ugh) so we're looking at go-live on the 24th instead of the 14th. cost is basically the same, maybe +$3k for overtime. Priya's team is crushing it tho. will ping u friday w/ more",
"facts": [
{
"name": "warehouse",
"any": [
"warehouse"
]
},
{
"name": "shelving",
"any": [
"shelving"
]
},
{
"name": "10 days late",
"any": [
"10 days",
"ten days"
]
},
{
"name": "go-live 24th",
"any": [
"24th"
]
},
{
"name": "instead of 14th",
"any": [
"14th"
]
},
{
"name": "$3k overtime",
"any": [
"$3k",
"$3000",
"$3 thousand",
"3000 dollars",
"3k"
]
},
{
"name": "overtime",
"any": [
"overtime"
]
},
{
"name": "Priya",
"any": [
"priya"
]
},
{
"name": "Friday",
"any": [
"friday"
]
}
],
"banned": [
"ugh",
"kinda",
"tho",
"u",
"w/",
"crushing it",
"ping",
"ok so"
],
"banned_re": [],
"no_caps": false,
"caps_ok": [],
"extra_numbers_ok": [],
"length_ratio": [
0.5,
3
],
"reference": "Quick update on the warehouse project: we are running behind schedule. The new shelving arrived 10 days late, so go-live is now planned for the 24th instead of the 14th. Overall cost is largely unchanged, with a possible increase of about $3,000 for overtime. Priya's team has been doing excellent work. I will follow up on Friday with more details."
}What each model answered
MiMo-V2.6 Distill 9B
✓ pass · score 98 (scripted checks 100, judges 9.7/10 from 3) · 2.52 s
Quick update on the warehouse project: we are behind schedule. The new shelving arrived approximately 10 days late, which puts go-live at the 24th instead of the 14th. Cost is essentially unchanged, with an estimated additional $3k for overtime. Priya's team is performing well. I will follow up this Friday with more details.
Judge medians: Tone match 10 · Faithfulness 9 · Naturalness 10
Scoring detail
{
"cleaned_words": 53,
"similarity_to_source": 0.495,
"facts_missing": [],
"banned_hits": [],
"new_numbers": [],
"length_ratio": 0.964,
"length_bounds": [
0.5,
3
],
"gate": 1,
"checks": {
"facts": 1,
"banned": 1,
"numbers": 1,
"changed": 1,
"length": 1
}
}Ornith-1.5 9B
✓ pass · score 98 (scripted checks 100, judges 9.7/10 from 3) · 3.03 s
Subject: Update on Warehouse Shelving and Revised Go-Live Date Hi, A quick update on the warehouse project: we are currently running behind schedule. The new shelving arrived approximately 10 days late, which has pushed the go-live date from the 14th to the 24th. The cost remains essentially the same, with an estimated additional $3,000 for overtime. Priya's team is performing well and making strong progress. I will follow up with you on Friday with further details. Best regards,
Judge medians: Tone match 10 · Faithfulness 9 · Naturalness 10
Scoring detail
{
"cleaned_words": 78,
"similarity_to_source": 0.343,
"facts_missing": [],
"banned_hits": [],
"new_numbers": [],
"length_ratio": 1.418,
"length_bounds": [
0.5,
3
],
"gate": 1,
"checks": {
"facts": 1,
"banned": 1,
"numbers": 1,
"changed": 1,
"length": 1
}
}Granite 4.2 8B
✓ pass · score 95 (scripted checks 100, judges 9.0/10 from 3) · 3.08 s
Subject: Warehouse Go‑Live Update Hi [Recipient], I wanted to provide a brief update on the warehouse project. Due to the late arrival of the new shelving (approximately 10 days later than scheduled), our planned go‑live date has been postponed from 14th to 24th. The overall cost remains essentially the same, with an estimated additional $3,000 for overtime. Priya’s team is performing well. I will send you a more detailed update on Friday. Best regards, [Your Name]
Judge medians: Tone match 10 · Faithfulness 9 · Naturalness 8
Scoring detail
{
"cleaned_words": 76,
"similarity_to_source": 0.318,
"facts_missing": [],
"banned_hits": [],
"new_numbers": [],
"length_ratio": 1.382,
"length_bounds": [
0.5,
3
],
"gate": 1,
"checks": {
"facts": 1,
"banned": 1,
"numbers": 1,
"changed": 1,
"length": 1
}
}Spark-X2.5 4B
✓ pass · score 88 (scripted checks 100, judges 7.7/10 from 3) · 1.92 s
Here is a clear and professional update regarding the warehouse project. We are currently experiencing a delay with the new shelving, which arrived 10 days late. Consequently, we have rescheduled the go-live date to the 24th instead of the 14th. The cost remains largely unchanged, with an estimated additional overtime cost of $3,000. However, Priya's team is performing exceptionally well. I will follow up with further details on Friday.
Judge medians: Tone match 9 · Faithfulness 8 · Naturalness 6
Scoring detail
{
"cleaned_words": 69,
"similarity_to_source": 0.304,
"facts_missing": [],
"banned_hits": [],
"new_numbers": [],
"length_ratio": 1.255,
"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 86 (scripted checks 95, judges 7.7/10 from 3) · 1.05 s
We are behind schedule. The new shelving arrived approximately ten days late, so the go‑live date has been moved to the 24th instead of the 14th. The cost remains essentially the same, with an additional roughly $3,000 for overtime. Priya’s team is performing well. I will send you an update on Friday.
Judge medians: Tone match 9 · Faithfulness 7 · Naturalness 7
Scoring detail
{
"cleaned_words": 52,
"similarity_to_source": 0.389,
"facts_missing": [
"warehouse"
],
"banned_hits": [],
"new_numbers": [],
"length_ratio": 0.945,
"length_bounds": [
0.5,
3
],
"gate": 1,
"checks": {
"facts": 0.8889,
"banned": 1,
"numbers": 1,
"changed": 1,
"length": 1
}
}Ling-3.0-tiny (7.9B-A1.3B)
✓ pass · score 95 (scripted checks 100, judges 9.0/10 from 3) · 0.95 s
Here is the rewritten message: Hi, I wanted to provide an update on the warehouse situation. The new shelving arrived 10 days later than expected, which has resulted in a go-live date of the 24th instead of the 14th. The cost remains essentially the same, with an additional $3,000 for overtime. Priya's team is doing an excellent job on this. I will send you a more detailed update on Friday.
Judge medians: Tone match 10 · Faithfulness 7 · Naturalness 10
Scoring detail
{
"cleaned_words": 65,
"similarity_to_source": 0.43,
"facts_missing": [],
"banned_hits": [],
"new_numbers": [],
"length_ratio": 1.182,
"length_bounds": [
0.5,
3
],
"gate": 1,
"checks": {
"facts": 1,
"banned": 1,
"numbers": 1,
"changed": 1,
"length": 1
}
}MiniCPM5 2B
✓ pass · score 95 (scripted checks 100, judges 9.0/10 from 3) · 1.16 s
Update on the warehouse situation: the new shelving arrived approximately 10 days late, so the go-live is now scheduled for the 24th instead of the 14th. The cost is essentially the same, with a possible additional $3k for overtime. Priya's team is performing well. I will follow up with you by Friday with more details.
Judge medians: Tone match 10 · Faithfulness 8 · Naturalness 9
Scoring detail
{
"cleaned_words": 55,
"similarity_to_source": 0.523,
"facts_missing": [],
"banned_hits": [],
"new_numbers": [],
"length_ratio": 1,
"length_bounds": [
0.5,
3
],
"gate": 1,
"checks": {
"facts": 1,
"banned": 1,
"numbers": 1,
"changed": 1,
"length": 1
}
}K2-Horizon 0.9B
✓ pass · score 82 (scripted checks 100, judges 6.3/10 from 3) · 0.43 s
Okay, quick update on the warehouse issue—we're behind. The new shelving arrived late, roughly 10 days late, which means we're considering go-live on the 24th instead of the 14th. Costs are essentially the same, with a possible $3k increase for overtime. Priya's team is doing well, so I'll follow up with you Friday.
Judge medians: Tone match 6 · Faithfulness 7 · Naturalness 6
Scoring detail
{
"cleaned_words": 53,
"similarity_to_source": 0.495,
"facts_missing": [],
"banned_hits": [],
"new_numbers": [],
"length_ratio": 0.964,
"length_bounds": [
0.5,
3
],
"gate": 1,
"checks": {
"facts": 1,
"banned": 1,
"numbers": 1,
"changed": 1,
"length": 1
}
}