Unified Memory

Benchmarks / Which Small AI Model Should You Run on a Mac? We Ran 312 Tests / Read images

vision-025

hard chartcounting

The prompt

1 image attached (not shown).

user

How many regions in this chart had more than 50 installs? Reply with only the number (no units or currency symbols).
What we expected
{
  "answer": 4,
  "accept": [],
  "type": "number"
}

What each model answered

LFM2.5-VL 3B

✓ pass · score 100 · 2.89 s

4
Scoring detail
{
  "type": "number",
  "parsed": 4
}

MiMo-V2.6 Distill 9B

✓ pass · score 100 · 2.02 s

4
Scoring detail
{
  "type": "number",
  "parsed": 4
}

MiniCPM-V 4.7 1B

✗ fail · score 0 · 1.14 s

5
Scoring detail
{
  "type": "number",
  "parsed": 5
}

← vision-024All Read images itemsvision-026 →