Decision-1.0-Lex-0.6B

Lex — Latin for law. Give your decisions a rule.

Open Decision Foundation Models

Turn operational context into structured decisions. Lex is a specialist fine-tuned from Kai to select actions, test conditions and apply scoring rubrics across four operational workflows.

Try Decision Studio · Decision collection.

A sharper specialist

78.15% accuracy — 1.55 points above Laya Typed Decisions.

Lex contains 571,909,635 parameters. Both models were evaluated on the original 2,000-decision English typed-decisions test split, with complete inputs and the official fine-tuned Laya checkpoint. Lex answers 31 more decisions correctly.

Observed accuracy gain; the paired 95% interval is −0.20 to +3.15 points. Laya leads on Noul and probability-quality metrics. Full evaluation.

Built for operational decisions

Supply the context, question and candidate descriptions at runtime. Choice returns candidate probabilities, Noul estimates whether a condition holds, and Score returns an ordered distribution and expected value. Applications choose how to act on those outputs.

Lex is evaluated as an English specialist on these four workflows. For broader multilingual tasks, explore the general Kai model.

128 mixed questions in 154 ms — 57% lower latency. Automatic typed scheduling accelerates the measured SystemOne runtime with the same weights. Paired local AMD measurements on a fixed workload. Latency and scaling.

Download for local inference

hf download vllm-sr/Decision-1.0-Lex-0.6B --local-dir Decision-1.0-Lex-0.6B

This repository contains model files, provenance and the Transformers loading code. Serving requires a compatible vLLM Semantic Router Decision runtime, distributed separately. Check its hardware support before serving. The runtime must support vllm-sr-decision format version 1 and the file map in config.json. For local inference without a server, see Use with 🤗 Transformers below.

Use with 🤗 Transformers

The repository includes its inference code, so stock Transformers can download and run the complete model locally with trust_remote_code=True. system_one takes and returns the same System One request and response bodies as the Decision runtime; nothing is generated.

pip install "transformers>=4.57" torch safetensors huggingface_hub

from transformers import AutoModel

model = AutoModel.from_pretrained("vllm-sr/Decision-1.0-Lex-0.6B", trust_remote_code=True)
response = model.system_one(
    state="The parcel arrived damaged. Please send a replacement today.",
    questions={
        "route": {
            "type": "choice",
            "instructions": "Which team should handle this request?",
            "criteria": {"delivery": "Damaged or missing parcels", "billing": "Payments and invoices"},
        },
        "urgent": {
            "type": "noul",
            "instructions": "Does the customer request action today?",
        },
    },
)
print(response["answers"]["route"]["choice"], response["answers"]["urgent"]["noul"])

pipeline("decision", model="vllm-sr/Decision-1.0-Lex-0.6B", trust_remote_code=True) accepts the same request body. A malformed question is answered with an invalid_question error. The model loads on the first GPU when one is visible, otherwise on the CPU (pass device="cpu" or device="cuda:0" to choose); weights and arithmetic are FP32. A complete question, its candidates and the state are limited to 1,024 tokens; if a question is longer, every question of the request is answered with a max_length_exceeded error and nothing is truncated.

Use

Replace the placeholder with a SystemOne-compatible endpoint configured to serve Decision-1.0-Lex-0.6B, and set DECISION_API_KEY to that endpoint's key.

pip install typesafe-sdk

import os
from typesafe_sdk import TypeSafeClient, Choice, Noul

client = TypeSafeClient(
    api_key=os.environ["DECISION_API_KEY"],
    base_url="https://your-decision-endpoint.example",
    model="Decision-1.0-Lex-0.6B",
)
questions = {
    "route": Choice(instructions="Which team should handle this request?",
                    criteria={"delivery": "Damaged or missing parcels", "billing": "Payments and invoices"}),
    "urgent": Noul(instructions="Does the customer request action today?"),
}
response = client.system_one(state="The parcel arrived damaged. Please send a replacement today.", questions=questions)
print(response.choices["route"].choice, response.nouls["urgent"].noul)

The same request with curl:

curl -X POST https://your-decision-endpoint.example/v1/systemone \
  -H "Authorization: Bearer $DECISION_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{
    "model": "Decision-1.0-Lex-0.6B",
    "state": "The parcel arrived damaged. Please send a replacement today.",
    "questions": {
      "route": {"type": "choice", "instructions": "Which team should handle this request?", "criteria": {"delivery": "Damaged or missing parcels", "billing": "Payments and invoices"}},
      "urgent": {"type": "noul", "instructions": "Does the customer request action today?"}
    }
  }'

Official Python SDK · HTTP API

Make it yours

One state. Many decisions. A compatible System One endpoint can submit typed questions and return results under their original question IDs. Request limits and batching depend on that deployment.

Use the SDK and curl examples with a compatible endpoint. Lex's specialization recipe and source data are documented in Methods and Training provenance.

The complete 1,024-token budget includes context, instructions, all candidates and special tokens. Overlength requests return an error. Native Choice and Score support 2–255 candidates or ordered levels; the System One and Studio interfaces use 2–10 Score levels.

Architecture

Three 22-layer bidirectional encoder paths share multilingual input embeddings, with separate interaction layers and candidate readouts for Choice, Noul and Score. Lex retains Kai's architecture and specializes its weights through supervised fine-tuning.

Architecture details · Training and methods

Built on Kai and Vela Encoder. Probabilities are not calibrated confidence; candidate order and task wording can affect outputs. Attribution and retained third-party terms · License scope.