Unified Memory

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

triage-012

medium inbox

The prompt

user

TASK: INBOX

From: Copperpot Kitchen <hello@copperpotkitchen.com>
To: dana.whitfield@inboxly.com
Date: Wed, 21 Oct 2026 12:00 -0400
Subject: Ends TONIGHT: 30% off Dutch ovens + our coziest soup recipes

Soup season is here!

THIS WEEK'S RECIPES
- Roasted squash and sage soup
- Weeknight chicken tortilla soup
- Slow-cooker white bean and kale stew

FLASH SALE - ENDS TONIGHT AT MIDNIGHT
Take 30% off every enameled Dutch oven with code SOUPSZN. Don't miss it!
Shop the sale: https://shop.copperpotkitchen.com/sale

You're getting this email because you signed up at copperpotkitchen.com.
Unsubscribe | Update preferences
Copperpot Kitchen, 22 Maple Row, Burlington, VT 05401

Reply with only a JSON object: {"label": "<urgent_action|reply_needed|fyi|newsletter|spam|phishing>"}
What we expected
{
  "label": "newsletter"
}

What each model answered

MiMo-V2.6 Distill 9B

✓ pass · score 100 · 0.74 s

{"label": "newsletter"}
Scoring detail
{
  "label": {
    "got": "newsletter",
    "expected": "newsletter",
    "correct": true
  }
}

Ornith-1.5 9B

✓ pass · score 100 · 0.68 s

{"label": "newsletter"}
Scoring detail
{
  "label": {
    "got": "newsletter",
    "expected": "newsletter",
    "correct": true
  }
}

Granite 4.2 8B

✓ pass · score 100 · 0.81 s

{"label": "newsletter"}
Scoring detail
{
  "label": {
    "got": "newsletter",
    "expected": "newsletter",
    "correct": true
  }
}

Spark-X2.5 4B

✓ pass · score 100 · 0.52 s

{"label": "newsletter"}
Scoring detail
{
  "label": {
    "got": "newsletter",
    "expected": "newsletter",
    "correct": true
  }
}

Mellum2.1 12B-A2.5B

✗ fail · score 0 · 0.32 s

{"label": "spam"}
Scoring detail
{
  "label": {
    "got": "spam",
    "expected": "newsletter",
    "correct": false
  }
}

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

✓ pass · score 100 · 0.23 s

{"label": "newsletter"}
Scoring detail
{
  "label": {
    "got": "newsletter",
    "expected": "newsletter",
    "correct": true
  }
}

MiniCPM5 2B

✗ fail · score 0 · 0.34 s

{"label": "spam"}
Scoring detail
{
  "label": {
    "got": "spam",
    "expected": "newsletter",
    "correct": false
  }
}

K2-Horizon 0.9B

✓ pass · score 100 · 0.17 s

{"label": "newsletter"}
Scoring detail
{
  "label": {
    "got": "newsletter",
    "expected": "newsletter",
    "correct": true
  }
}

← triage-011All Sort and triage itemstriage-013 →