Instructions to use yothinS/Qwen3-4B-ORPO-Advertisement with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yothinS/Qwen3-4B-ORPO-Advertisement with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="yothinS/Qwen3-4B-ORPO-Advertisement")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("yothinS/Qwen3-4B-ORPO-Advertisement")
model = AutoModelForCausalLM.from_pretrained("yothinS/Qwen3-4B-ORPO-Advertisement", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yothinS/Qwen3-4B-ORPO-Advertisement with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yothinS/Qwen3-4B-ORPO-Advertisement"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yothinS/Qwen3-4B-ORPO-Advertisement",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yothinS/Qwen3-4B-ORPO-Advertisement

SGLang

How to use yothinS/Qwen3-4B-ORPO-Advertisement with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "yothinS/Qwen3-4B-ORPO-Advertisement" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yothinS/Qwen3-4B-ORPO-Advertisement",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "yothinS/Qwen3-4B-ORPO-Advertisement" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yothinS/Qwen3-4B-ORPO-Advertisement",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use yothinS/Qwen3-4B-ORPO-Advertisement with Docker Model Runner:
```
docker model run hf.co/yothinS/Qwen3-4B-ORPO-Advertisement
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Qwen3-4B-ORPO-Advertisement

A specialized language model for Advertisement Strategy, Advertisement Analysis, and Performance Marketing, fine-tuned from Qwen3-4B using Odds Ratio Preference Optimization (ORPO).

Model Overview

yothinS/Qwen3-4B-ORPO-Advertisement is a fine-tuned language model designed to assist with advertising strategy, campaign performance analysis, and data-driven marketing decisions.

Built on top of Qwen/Qwen3-4B, this model uses Odds Ratio Preference Optimization (ORPO) to align its responses with preferred advertising-related answers using a specialized preference dataset.

The goal is to develop a more domain-focused assistant that can help marketers interpret advertising metrics, diagnose campaign performance, develop marketing strategies, and make better-informed optimization decisions.

Key Features

Advertisement Strategy: Develop advertising strategies based on campaign objectives, target audiences, budgets, and business goals.
Advertisement Analysis: Interpret campaign metrics and identify potential performance issues.
Performance Marketing: Analyze CTR, CPC, CVR, CPA, ROAS, and other advertising KPIs.
Audience Strategy: Support audience segmentation, targeting, customer acquisition, and retention planning.
Conversion Optimization: Identify potential improvements across advertising funnels and conversion journeys.
Creative & Messaging: Assist with advertising hooks, headlines, copywriting, calls to action, and creative angles.
Campaign Optimization: Recommend data-informed experiments and optimization opportunities.
Marketing Decision Support: Explain trade-offs between campaign performance, customer acquisition costs, revenue, and profitability.

These are the model's intended capabilities, not a guarantee that every response will be accurate or optimal.

Model Details

Property	Description
Model Name	Qwen3-4B-ORPO-Advertisement
Hugging Face Repository	`yothinS/Qwen3-4B-ORPO-Advertisement`
Base Model	Qwen/Qwen3-4B
Model Architecture	Qwen3
Approximate Parameters	4 billion
Fine-Tuning Method	Odds Ratio Preference Optimization (ORPO)
Primary Domain	Advertisement & Performance Marketing
Intended Tasks	Text generation, marketing analysis, and strategic decision support
Languages	English and Thai are intended for use; actual performance may vary
Training Dataset	yothinS/Advertisement_specialist

Training Method

Base Model

The model is derived from Qwen3-4B, a pretrained language model developed by the Qwen team.

Original model: https://huggingface.co/Qwen/Qwen3-4B

Fine-Tuning Approach: ORPO

Odds Ratio Preference Optimization (ORPO) is a preference-alignment method that combines supervised learning with an odds-ratio-based preference objective.

In preference fine-tuning, the model learns to favor a preferred response (chosen) over a less-preferred response (rejected) for a given prompt.

For this project, ORPO is used to encourage responses that better reflect the quality and reasoning patterns represented in the advertising-specialist preference dataset.

The intended outcome is a model that is more consistent with the preferred responses for advertising-related tasks.

Training Dataset

The model was fine-tuned using the following dataset:

Advertisement Specialist Dataset

The dataset provides advertising-related preference examples intended to guide the model toward more useful responses in advertisement strategy and analysis.

Dataset repository: yothinS/Advertisement_specialist

The dataset's published card should be consulted for its exact record count, schema, data-generation methodology, and licensing information.

Training Configuration

The following details should be updated to match the actual training run.

Parameter	Value
Base Model	Qwen/Qwen3-4B
Fine-Tuning Method	ORPO
Training Framework	Unsloth
Training Dataset	Advertisement Specialist Dataset
Training Records	877 record
Epochs	2
Learning Rate	5e-5
Maximum Sequence Length	1024
LoRA / QLoRA	QLoRA
Hardware	T4 16GB VRAM

Intended Use

This model is intended to support marketing professionals, advertisers, performance marketers, business owners, marketing analysts, and AI developers.

Recommended use cases include:

Advertising Strategy Development
- Develop campaign strategies around business objectives.
- Select target audiences and appropriate campaign approaches.
- Plan customer acquisition and retention initiatives.
Advertising Performance Analysis
- Interpret CTR, CPC, CVR, CPA, and ROAS.
- Diagnose potential campaign inefficiencies.
- Identify metrics that require further investigation.
Budget and Campaign Optimization
- Compare campaign performance and budget allocation options.
- Evaluate trade-offs between acquisition cost and return.
- Suggest hypotheses for controlled experiments.
Creative and Copywriting
- Generate advertising hooks, headlines, and calls to action.
- Explore alternative creative angles and value propositions.
- Improve alignment between messaging and target audiences.
Marketing Analytics and Decision Support
- Explain campaign results in business terms.
- Identify missing data needed for reliable decisions.
- Distinguish observed performance from potential causes.

Example Prompts

Example 1: ROAS Analysis

Prompt:

An advertising campaign generated ฿120,000 in attributed revenue with an advertising spend of ฿30,000. Calculate ROAS and explain what the result means.

Expected task:

Calculate ROAS as revenue divided by advertising spend, yielding 4.0x. Explain that each ฿1 of advertising spend generated ฿4 in attributed revenue, while clarifying that ROAS alone does not establish profitability.

Example 2: CTR Diagnosis

Prompt:

A campaign has a high CTR but a low conversion rate. What should I investigate?

Expected task:

Investigate audience-message alignment, landing-page relevance, offer clarity, page speed, conversion tracking, and potential friction in the conversion journey. Treat these as hypotheses to validate rather than confirmed causes.

Example 3: Campaign Strategy

Prompt:

I want to improve customer acquisition while controlling CPA. What advertising experiments should I consider?

Expected task:

Recommend structured experiments involving audience segments, creative variations, landing pages, and bidding or budget strategies, with appropriate success metrics and safeguards against misleading comparisons.

These examples illustrate intended use. They are not verified outputs from a model evaluation.

How to Use

Install the required libraries:

pip install -U transformers accelerate torch

Load the model with Hugging Face Transformers:

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "yothinS/Qwen3-4B-ORPO-Advertisement"

tokenizer = AutoTokenizer.from_pretrained(model_id)

model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

messages = [
    {
        "role": "user",
        "content": (
            "Analyze this advertising campaign: "
            "Spend = 10000 THB, Revenue = 40000 THB. "
            "Calculate ROAS and explain its limitations."
        ),
    }
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=False,
)

inputs = tokenizer(text, return_tensors="pt").to(model.device)

with torch.no_grad():
    outputs = model.generate(
        **inputs,
        max_new_tokens=512,
        do_sample=False,
    )

generated_tokens = outputs[0][inputs["input_ids"].shape[1]:]

response = tokenizer.decode(
    generated_tokens,
    skip_special_tokens=True,
)

print(response)

Note: This example assumes the repository contains the complete model weights and tokenizer files required by Transformers. If the repository contains only a LoRA adapter, load the compatible base model and adapter using PEFT instead. The enable_thinking argument should be used only if supported by the tokenizer's chat template.

Evaluation

A reliable evaluation should test the model on held-out advertising problems that were not included in training.

Suggested evaluation dimensions:

Metric Accuracy: Correct calculations and interpretations of CTR, CPC, CVR, CPA, and ROAS.
Analytical Quality: Ability to distinguish symptoms, hypotheses, and evidence-supported conclusions.
Strategic Relevance: Practicality and alignment of recommendations with stated business objectives.
Financial Reasoning: Awareness of margins, contribution profit, attribution limitations, and profitability.
Instruction Following: Ability to follow requested output formats and constraints.
Thai and English Performance: Quality and consistency across both intended languages.

No benchmark scores or comparative performance claims are reported in this model card. Evaluation results should be added after testing.

Limitations and Risks

Not a substitute for professional judgment: Recommendations should be reviewed before implementation.
Potentially incorrect calculations: Verify financial figures and KPI calculations independently.
Attribution limitations: Attributed revenue and ROAS do not necessarily represent incremental revenue or net profit.
Incomplete context: Recommendations may be unreliable when campaign objectives, margins, conversion tracking, or audience details are missing.
Synthetic or imperfect preferences: If the training dataset contains synthetic, inconsistent, or biased examples, these characteristics may influence the model.
No guaranteed live data access: The model should not be assumed to have access to current advertising platform metrics, campaign dashboards, or market conditions.
Potential overgeneralization: Advertising strategies that work in one industry, market, or audience segment may not transfer to another.
Language variation: Thai and English outputs may differ in quality depending on the task and available training examples.

For important marketing or financial decisions, validate recommendations against real campaign data and controlled experiments.

Ethical Considerations

This model should be used for responsible advertising practices. Users should respect privacy, applicable data protection laws, advertising policies, and consumer rights.

Avoid using model-generated recommendations to create deceptive claims, exploit vulnerable audiences, discriminate unfairly, or violate advertising platform policies.

License

The base model, Qwen/Qwen3-4B, is published under the Apache-2.0 license. This model card provisionally identifies Apache-2.0 for the fine-tuned model; verify that the training dataset's license, redistribution conditions, and all applicable third-party terms permit this designation before publishing.

Base model license: https://huggingface.co/Qwen/Qwen3-4B

Dataset: https://huggingface.co/datasets/yothinS/Advertisement_specialist

Acknowledgments

Qwen Team — for the Qwen3-4B base model.
Hugging Face — for the model and dataset hosting ecosystem.
Advertisement Specialist Dataset — for the advertising-domain preference data used in this project.