d1 is a 3B decision model from Liquid AI, built on LFM2.5-VL-3B. Give it a state (text, JSON, images, or a mix) and a set of typed questions, and it returns a calibrated probability for every answer in a single forward pass. It isn’t a chat model and doesn’t generate text.

It’s compatible with the Jev and System One APIs, so it works through Ollama’s /v1/systemone endpoint and ollama run --ask.

ollama pull d1

Highlights

• Small and accurate: Liquid AI reports 48.57 on Decision Index 0.2.1, the best score for a decision model under 10B, ahead of the 35B-A3B Decider (47.11).
• Multimodal: images and text in the same state.
• One forward pass: every option of every question is scored together, with no output tokens.
• Multilingual: Arabic, Chinese, English, French, German, Hindi, Indonesian, Italian, Japanese, Korean, Polish, Portuguese, Russian, Spanish, Thai, and Vietnamese.

What you can build

CLI

ollama run d1 --ask "Is a refund being requested?" < ticket.txt

ollama run d1 --ask "Which team should handle this?" \
  --choice billing="Payments and refunds" \
  --choice technical="Outages and errors" < ticket.txt

ollama run d1 --ask "How urgent is this?" \
  --level "Routine" \
  --level "Soon" \
  --level "Immediate" < ticket.txt

The state can also follow the model name instead of coming from stdin.

API

Decision models use Ollama’s /v1/systemone endpoint, documented here. Put the text you want judged in state and your questions in questions. Ollama builds d1’s prompt for you.

Every question has a type, instructions, and usually criteria:

confidence runs from 0 to 1 and shows how strongly the probabilities favor one answer. It isn’t the chance that the answer is right.

cURL

curl http://localhost:11434/v1/systemone -d '{
  "model": "d1",
  "state": "Our app crashes on launch for every user since this morning'\''s update. Please fix it now.",
  "questions": {
    "urgent": {"type": "noul", "instructions": "Is this support request urgent?"},
    "team": {
      "type": "choice",
      "instructions": "Which team should handle this request?",
      "criteria": {
        "billing": "Payments, invoices, and refunds",
        "technical": "Outages, errors, and configuration",
        "sales": "Plans and upgrades"
      }
    },
    "severity": {
      "type": "score",
      "instructions": "How severe is the customer impact?",
      "criteria": ["No impact", "Minor", "Major", "Critical"]
    }
  }
}'

Response:

{
  "model": "d1",
  "answers": {
    "urgent": {"type": "noul", "noul": 0.981},
    "team": {
      "type": "choice",
      "choice": "technical",
      "probabilities": {"billing": 0.012, "technical": 0.973, "sales": 0.015},
      "confidence": 0.96
    },
    "severity": {
      "type": "score",
      "score": 2.582,
      "legend": {"0": "No impact", "1": "Minor", "2": "Major", "3": "Critical"},
      "probabilities": {"0": 0.015, "1": 0.012, "2": 0.35, "3": 0.623},
      "confidence": 0.582
    }
  },
  "usage": {"input_tokens": 175, "output_tokens": 0}
}

Python

Use TypeSafe’s official Python SDK and point it at Ollama. The SDK requires an API key, but Ollama ignores it, so any value works.

pip install typesafe-sdk
export TYPESAFE_BASE_URL=http://localhost:11434
export TYPESAFE_API_KEY=ollama
export TYPESAFE_DEFAULT_MODEL=d1

from typesafe_sdk import Choice, Noul, Score, TypeSafeClient

questions = {
    "urgent": Noul(instructions="Is this support request urgent?"),
    "team": Choice(
        instructions="Which team should handle this request?",
        criteria={"billing": "Payments, invoices, and refunds", "technical": "Outages, errors, and configuration", "sales": "Plans and upgrades"},
    ),
    "severity": Score(instructions="How severe is the customer impact?", criteria=["No impact", "Minor", "Major", "Critical"]),
}

with TypeSafeClient(timeout=120) as client:
    result = client.system_one(state="Our app crashes on launch for every user since this morning's update. Please fix it now.", questions=questions)

print(result.choices["team"].choice)  # technical

Images

Images are shared by all questions:

curl http://localhost:11434/v1/systemone -d '{
  "model": "d1",
  "state": "Photo from the returns desk.",
  "images": ["<base64-encoded image>"],
  "questions": {
    "damaged": {"type": "noul", "instructions": "Is the product visibly damaged?"},
    "item": {
      "type": "choice",
      "instructions": "What is in the photo?",
      "criteria": {"shoes": null, "electronics": null, "clothing": null, "other": null}
    }
  }
}'

Benchmarks

Results reported by Liquid AI. See the model card and blog post for details.

License

LFM Open License v1.0