wozniakos10

Qwen3.5-4B-typed-decisions-merged

README

License: apache-2.0

Question types

Table with columns: Type, Answers, Output
Type	Answers	Output
`noul`	A yes/no question	Probability that the answer is yes
`choice`	One option out of named criteria	Probability per option
`score`	A position on an ordered scale	Probability per scale level

Usage

Install Unsloth, then:

python

from unsloth import FastDecisionModel model, tokenizer = FastDecisionModel.from_pretrained("wozniakos10/Qwen3.5-4B-typed-decisions-merged", load_in_4bit = True)FastDecisionModel.for_inference(model) answers = FastDecisionModel.predict(    model,    tokenizer,    "Hi, I was charged twice for invoice #4411. Please refund the duplicate today.",    {        "team": {            "type": "choice",            "instructions": "Which team should handle this?",            "criteria": {                "billing": "invoices, payments, refunds",                "technical": "bugs, outages, errors",                "sales": "pricing, new plans",            },        },        "refund": {"type": "noul", "instructions": "Does the customer ask for a refund?"},        "urgency": {            "type": "score",            "instructions": "How urgent is this?",            "criteria": ["not urgent", "soon", "today"],        },    },)for name, result in answers.items():    print(name, result["answer"], result["probabilities"])

Intended use

Routing, triage, tagging, filtering and other places where code needs a structured judgment with a probability instead of free text. Questions and criteria are supplied at inference time, so no retraining is needed for new labels.

Training

Trained with DecisionTrainer on the all config, train split of LocalLLaMA/typed-decisions. A part of the training rows was held out with FastDecisionModel.split_holdout. The output probabilities have not been temperature-calibrated.

Table with columns: Hyperparameter, Value
Hyperparameter	Value
Training steps	240
Epochs	1.71
Learning rate (LoRA adapters)	2e-5
Learning rate (decision head)	1e-4
LR scheduler	cosine
Warmup steps	10
Weight decay	0.01
Per-device batch size	2
Gradient accumulation steps	4

Table with columns: Metric, Value
Metric	Value
Final training loss	0.7875
Held-out loss at last evaluation	0.8747

Architecture

Table with columns: Component, Value
Component	Value
Base model	`unsloth/Qwen3.5-4B`
Decision layout	`clef`
Decision head hidden_size	2560
Decision head width	512
Decision head routing_layers	2
Decision head layers	4
Decision head heads	8
Decision head feedforward

Evaluation

Results of FastDecisionModel.evaluate on the all config, test split of LocalLLaMA/typed-decisions.

Table with columns: Metric, Value
Metric	Value
accuracy	0.7715
ece	0.1381
loss	0.8664
record_accuracy	0.345

Limitations

Probabilities are not temperature-calibrated, so validate on your own data before relying on thresholds.
Trained on English data. Other languages are untested.
Inputs longer than 2048 tokens are outside the trained range.
Outputs are judgments, not facts. Keep a human in the loop for high-stakes decisions.

License

Apache 2.0, inherited from the base model. Check the terms of LocalLLaMA/typed-decisions before commercial use.

Acknowledgements

Trained with the Unsloth decision model guide.

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

wozniakos10

Model Tree

Base

unsloth/Qwen3.5-4B

Fine-tuned

this model

Input Modalities

TextImageVideo

Output Modalities

Text

Supported Functionality