Instructions to use Alvaropar/FinQwen3-8B-adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Alvaropar/FinQwen3-8B-adapter with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B")
model = PeftModel.from_pretrained(base_model, "Alvaropar/FinQwen3-8B-adapter")

Transformers

How to use Alvaropar/FinQwen3-8B-adapter with Transformers:

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

pipe = pipeline("text-generation", model="Alvaropar/FinQwen3-8B-adapter")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Alvaropar/FinQwen3-8B-adapter", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Alvaropar/FinQwen3-8B-adapter with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Alvaropar/FinQwen3-8B-adapter"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Alvaropar/FinQwen3-8B-adapter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Alvaropar/FinQwen3-8B-adapter

SGLang

How to use Alvaropar/FinQwen3-8B-adapter 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 "Alvaropar/FinQwen3-8B-adapter" \
    --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": "Alvaropar/FinQwen3-8B-adapter",
		"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 "Alvaropar/FinQwen3-8B-adapter" \
        --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": "Alvaropar/FinQwen3-8B-adapter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Alvaropar/FinQwen3-8B-adapter with Docker Model Runner:
```
docker model run hf.co/Alvaropar/FinQwen3-8B-adapter
```

Model Card for Qwen3-8B Finetuned (Financial Sentiment - LoRA)

Model Details

Model Description

This model is a parameter-efficient fine-tuned version of Qwen3-8B, adapted using LoRA (Low-Rank Adaptation) for financial sentiment analysis.

It is trained as an instruction-following text generation model that classifies financial sentences into:

positive
negative
neutral

The model generates a single-word output, making it suitable for lightweight classification pipelines while retaining generative flexibility.

Developed by: Authors of the referenced paper
Model type: Causal Language Model (decoder-only transformer)
Language(s): English (financial domain)
License: Same as base model (Qwen license – verify before use)
Finetuned from model: Qwen/Qwen3-8B

Model Sources

Base Model: Qwen/Qwen3-8B
Paper: https://arxiv.org/pdf/2512.00946

Uses

Direct Use

This model is intended for:

Financial sentiment classification
Text generation with constrained outputs
Research on instruction-tuned LLM classifiers

Example applications:

Market sentiment analysis
News sentiment tagging
Signal generation for trading pipelines

Downstream Use

Integration into trading or analytics systems
Combination with time-series models (e.g., LSTM, TFT)
Use in RAG pipelines for financial reasoning

Out-of-Scope Use

High-stakes financial decision-making without validation
Legal or medical analysis
Tasks requiring guaranteed factual correctness
General sentiment analysis outside financial domain (may degrade)

Prompt Format Requirement

This model is highly sensitive to prompt structure.

It was trained using a fixed instruction template, and performance depends heavily on using the same format at inference time.

Required Prompt

You are an AI language model trained to detect the sentiment of each sentence for finance trends.
Analyze the following sentence and determine if the sentiment is: positive or negative or neutral.
Return only a single word, either positive or negative or neutral.

<your input text>

Downloads last month: 2

Model tree for Alvaropar/FinQwen3-8B-adapter

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Adapter

(1941)

this model

Paper for Alvaropar/FinQwen3-8B-adapter

Fine-tuning of lightweight large language models for sentiment classification on heterogeneous financial textual data

Paper • 2512.00946 • Published Nov 30, 2025