Instructions to use kabesaml/regllm-qwen25-7b-banking-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kabesaml/regllm-qwen25-7b-banking-lora with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-7B-Instruct")
model = PeftModel.from_pretrained(base_model, "kabesaml/regllm-qwen25-7b-banking-lora")

Transformers

How to use kabesaml/regllm-qwen25-7b-banking-lora with Transformers:

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

pipe = pipeline("text-generation", model="kabesaml/regllm-qwen25-7b-banking-lora")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("kabesaml/regllm-qwen25-7b-banking-lora", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use kabesaml/regllm-qwen25-7b-banking-lora with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kabesaml/regllm-qwen25-7b-banking-lora"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kabesaml/regllm-qwen25-7b-banking-lora",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kabesaml/regllm-qwen25-7b-banking-lora

SGLang

How to use kabesaml/regllm-qwen25-7b-banking-lora 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 "kabesaml/regllm-qwen25-7b-banking-lora" \
    --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": "kabesaml/regllm-qwen25-7b-banking-lora",
		"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 "kabesaml/regllm-qwen25-7b-banking-lora" \
        --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": "kabesaml/regllm-qwen25-7b-banking-lora",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kabesaml/regllm-qwen25-7b-banking-lora with Docker Model Runner:
```
docker model run hf.co/kabesaml/regllm-qwen25-7b-banking-lora
```

RegLLM — Banking & Credit Risk Expert (LoRA adapter)

A LoRA adapter fine-tuned on top of Qwen/Qwen2.5-7B-Instruct for banking regulation and credit risk expertise.

What it does

This model specialises in:

Spanish banking regulation: EBA Guidelines, CRR/CRD, Basel accords
Credit risk methodology: PD/LGD/EAD estimation, IRB models, IFRS 9, stress testing
SQL methodology review: validates credit risk SQL code against regulatory standards
Spanish bank financials: Santander, BBVA, CaixaBank, Sabadell, Kutxabank (2022–2023)

Training Details

Parameter	Value
Base model	Qwen/Qwen2.5-7B-Instruct
Method	LoRA (SFT)
LoRA rank	32
LoRA alpha	64
Training examples	~163 (SQL + Banking Q&A + Regulation)
Adapter size	~309MB
Run	2026-02-22

Quick Start

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch

base_model = "Qwen/Qwen2.5-7B-Instruct"
adapter = "kabesaml/regllm-qwen25-7b-banking-lora"

tokenizer = AutoTokenizer.from_pretrained(base_model)
model = AutoModelForCausalLM.from_pretrained(
    base_model,
    torch_dtype=torch.float16,
    device_map="auto"
)
model = PeftModel.from_pretrained(model, adapter)

messages = [
    {"role": "system", "content": "Eres un asistente experto en regulación bancaria y el sector bancario español. Responde con datos precisos y cita la normativa cuando sea posible."},
    {"role": "user", "content": "¿Qué es la tasa de impago bajo el estándar CRR Art. 178?"}
]

text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512, temperature=0.7)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

System Prompt

Eres un asistente experto en regulación bancaria y el sector bancario español.
Responde con datos precisos y cita la normativa cuando sea posible.

Limitations

Trained on a relatively small dataset (~163 examples); best used as a specialised augmentation over the base model
Primarily focused on Spanish/EU banking regulation
Not a substitute for professional regulatory advice

Downloads last month: 10

Model tree for kabesaml/regllm-qwen25-7b-banking-lora

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-7B-Instruct

Adapter

(2629)

this model

kabesaml
/

regllm-qwen25-7b-banking-lora