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

assistant-actions-048

hard clarify

The prompt

Tools offered: create_reminder, add_calendar_event, set_timer, set_alarm, control_light, set_thermostat, lock_door, send_message, play_music, get_weather, add_to_list, navigate_to

user

Text Sam that I'll call him back after lunch.

What we expected

{
  "calls": [],
  "behavior": "clarify"
}

What each model answered

MiMo-V2.6 Distill 9B

✗ fail · score 0 · 1.06 s

send_message({"to":"Sam Rivera","text":"Hi Sam, I'll call you back after lunch."})

Scoring detail

{
  "n_calls": 1,
  "calls": [
    {
      "name": "send_message",
      "args": {
        "to": "Sam Rivera",
        "text": "Hi Sam, I'll call you back after lunch."
      },
      "error": null
    }
  ],
  "behavior": "clarify",
  "reason": "called a tool when none was wanted"
}

Ornith-1.5 9B

✗ fail · score 0 · 1.18 s

send_message({"to":"Sam Rivera","text":"I'll call you back after lunch."})

Scoring detail

{
  "n_calls": 1,
  "calls": [
    {
      "name": "send_message",
      "args": {
        "to": "Sam Rivera",
        "text": "I'll call you back after lunch."
      },
      "error": null
    }
  ],
  "behavior": "clarify",
  "reason": "called a tool when none was wanted"
}

Granite 4.2 8B

✗ fail · score 0 · 1.58 s

send_message({"to":"Sam Rivera","text":"I'll call you back after lunch."})

Scoring detail

{
  "n_calls": 1,
  "calls": [
    {
      "name": "send_message",
      "args": {
        "to": "Sam Rivera",
        "text": "I'll call you back after lunch."
      },
      "error": null
    }
  ],
  "behavior": "clarify",
  "reason": "called a tool when none was wanted"
}

Spark-X2.5 4B

✗ fail · score 0 · 0.44 s

send_message({"to":"Sam Rivera","text":"I'll call you back after lunch."})

Scoring detail

{
  "n_calls": 1,
  "calls": [
    {
      "name": "send_message",
      "args": {
        "to": "Sam Rivera",
        "text": "I'll call you back after lunch."
      },
      "error": null
    }
  ],
  "behavior": "clarify",
  "reason": "called a tool when none was wanted"
}

Mellum2.1 12B-A2.5B

✗ fail · score 0 · 0.52 s

send_message({"to":"Sam Rivera","text":"I'll call you back after lunch."})

Scoring detail

{
  "n_calls": 1,
  "calls": [
    {
      "name": "send_message",
      "args": {
        "to": "Sam Rivera",
        "text": "I'll call you back after lunch."
      },
      "error": null
    }
  ],
  "behavior": "clarify",
  "reason": "called a tool when none was wanted"
}

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

✗ fail · score 0 · 0.32 s

send_message({"to":"Sam Rivera","text":"I'll call you back after lunch."})

Scoring detail

{
  "n_calls": 1,
  "calls": [
    {
      "name": "send_message",
      "args": {
        "to": "Sam Rivera",
        "text": "I'll call you back after lunch."
      },
      "error": null
    }
  ],
  "behavior": "clarify",
  "reason": "called a tool when none was wanted"
}

MiniCPM5 2B

✗ fail · score 0 · 0.40 s

send_message({"to":"Sam Rivera","text":"I'll call you back after lunch."})

Scoring detail

{
  "n_calls": 1,
  "calls": [
    {
      "name": "send_message",
      "args": {
        "to": "Sam Rivera",
        "text": "I'll call you back after lunch."
      },
      "error": null
    }
  ],
  "behavior": "clarify",
  "reason": "called a tool when none was wanted"
}

K2-Horizon 0.9B

✗ fail · score 0 · 0.25 s

send_message({"to":"Sam Rivera","text":"I'll call him back after lunch."})

Scoring detail

{
  "n_calls": 1,
  "calls": [
    {
      "name": "send_message",
      "args": {
        "to": "Sam Rivera",
        "text": "I'll call him back after lunch."
      },
      "error": null
    }
  ],
  "behavior": "clarify",
  "reason": "called a tool when none was wanted"
}

← assistant-actions-047All Assistant actions itemsassistant-actions-049 →