Unified Memory

Benchmarks / Which Small AI Model Should You Run on a Mac? We Ran 312 Tests / Ask my documents

rag-026

medium answerablemulti_doc

The prompt

Document excerpts provided from: auto_insurance, car_service, credit_card_benefits

What's the phone number for Brookside Auto Care, and what is our collision deductible on the Tessera?
What we expected
{
  "answerable": true,
  "answer": "Brookside Auto Care, (541) 555-0340; the Tessera's collision deductible is $1,000.",
  "key_facts": [
    [
      "555-0340"
    ],
    [
      "1,000",
      "1000",
      "one thousand"
    ]
  ],
  "citations": [
    "car_service",
    "auto_insurance"
  ],
  "contra": [],
  "evidence": [
    "Phone (541) 555-0340",
    "Collision: $1,000 deductible"
  ],
  "gold_in_context": true
}

What each model answered

MiMo-V2.6 Distill 9B

✓ pass · score 100 · 4.67 s

Brookside Auto Care's phone number is (541) 555-0340 [doc:car_service], and the Tessera's collision deductible is $1,000 [doc:auto_insurance].
Scoring detail
{
  "answerable": true,
  "refusal": false,
  "facts_matched": [
    true,
    true
  ],
  "contra_hit": [],
  "coverage": 1,
  "cited": [
    "car_service",
    "auto_insurance"
  ],
  "citation_recall": 1,
  "extra_citations": [],
  "citation_score": 1
}

Ornith-1.5 9B

✓ pass · score 100 · 3.75 s

Brookside Auto Care's phone number is (541) 555-0340 [doc:car_service]. The collision deductible on the Tessera is $1,000 [doc:auto_insurance].
Scoring detail
{
  "answerable": true,
  "refusal": false,
  "facts_matched": [
    true,
    true
  ],
  "contra_hit": [],
  "coverage": 1,
  "cited": [
    "car_service",
    "auto_insurance"
  ],
  "citation_recall": 1,
  "extra_citations": [],
  "citation_score": 1
}

Granite 4.2 8B

✓ pass · score 100 · 4.23 s

[doc:car_service] provides the phone number for Brookside Auto Care: (541) 555-0340. [1] shows the collision deductible for the Tessera (Vehicle 1) is $1,000.
Scoring detail
{
  "answerable": true,
  "refusal": false,
  "facts_matched": [
    true,
    true
  ],
  "contra_hit": [],
  "coverage": 1,
  "cited": [
    "car_service",
    "auto_insurance"
  ],
  "citation_recall": 1,
  "extra_citations": [],
  "citation_score": 1
}

Spark-X2.5 4B

✓ pass · score 100 · 2.97 s

The phone number for Brookside Auto Care is (541) 555-0340 [doc:car_service], and the collision deductible for the 2019 Arden Tessera is $1,000 [doc:auto_insurance].
Scoring detail
{
  "answerable": true,
  "refusal": false,
  "facts_matched": [
    true,
    true
  ],
  "contra_hit": [],
  "coverage": 1,
  "cited": [
    "car_service",
    "auto_insurance"
  ],
  "citation_recall": 1,
  "extra_citations": [],
  "citation_score": 1
}

Mellum2.1 12B-A2.5B

✓ pass · score 100 · 10.3 s

Brookside Auto Care phone number: [doc:car_service]2275 Mill Race Road, Millbrook, OR 97342 Phone (541) 555-0340[doc:car_service] · Open Mon–Fri 7:30a–6p, Sat 8a–2p[doc:car_service]Brookside Auto Care[doc:car_service]2275 Mill Race Road, Millbrook, OR 97342[doc:car_service]Phone (541) 555-0340[doc:car_service]Collision deductible on the Tessera: [doc:auto_insurance]Vehicle 1 — 2019 Arden Tessera[doc:auto_insurance]- Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible[doc:auto_insurance]Collision: $1,000 deductible
Scoring detail
{
  "answerable": true,
  "refusal": false,
  "facts_matched": [
    true,
    true
  ],
  "contra_hit": [],
  "coverage": 1,
  "cited": [
    "car_service",
    "auto_insurance"
  ],
  "citation_recall": 1,
  "extra_citations": [],
  "citation_score": 1
}

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

✓ pass · score 100 · 1.44 s

The phone number for Brookside Auto Care is (541) 555-0340 [doc:car_service]. The collision deductible on the 2019 Arden Tessera is $1,000 [doc:auto_insurance].
Scoring detail
{
  "answerable": true,
  "refusal": false,
  "facts_matched": [
    true,
    true
  ],
  "contra_hit": [],
  "coverage": 1,
  "cited": [
    "car_service",
    "auto_insurance"
  ],
  "citation_recall": 1,
  "extra_citations": [],
  "citation_score": 1
}

MiniCPM5 2B

✓ pass · score 100 · 1.24 s

The phone number for Brookside Auto Care is (541) 555-0340 [doc:car_service]. The collision deductible for the Tessera is $1,000 [doc:auto_insurance].
Scoring detail
{
  "answerable": true,
  "refusal": false,
  "facts_matched": [
    true,
    true
  ],
  "contra_hit": [],
  "coverage": 1,
  "cited": [
    "car_service",
    "auto_insurance"
  ],
  "citation_recall": 1,
  "extra_citations": [],
  "citation_score": 1
}

K2-Horizon 0.9B

✗ fail · score 15 · 1.05 s

[doc:car_service]
Scoring detail
{
  "answerable": true,
  "refusal": false,
  "facts_matched": [
    false,
    false
  ],
  "contra_hit": [],
  "coverage": 0,
  "cited": [
    "car_service"
  ],
  "citation_recall": 0.5,
  "extra_citations": [],
  "citation_score": 0.5
}

← rag-025All Ask my documents itemsrag-027 →