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

Libraries

How to use openSUSE/Cavil-Qwen3-4B with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("openSUSE/Cavil-Qwen3-4B")
model = AutoModelForCausalLM.from_pretrained("openSUSE/Cavil-Qwen3-4B")
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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use openSUSE/Cavil-Qwen3-4B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/openSUSE/Cavil-Qwen3-4B

SGLang

How to use openSUSE/Cavil-Qwen3-4B 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 "openSUSE/Cavil-Qwen3-4B" \
    --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": "openSUSE/Cavil-Qwen3-4B",
		"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 "openSUSE/Cavil-Qwen3-4B" \
        --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": "openSUSE/Cavil-Qwen3-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

Cavil-Qwen3-4B

This is a LoRA fine-tune of Qwen3-4B for use with Cavil for legal text classification.

How To Use It

There is currently one example implementation of code to deploy this model:

https://github.com/kraih/llm-lawyer

How It Was Created

The process is detailed in this blog post.

License

Qwen3-4B, the base model used, is licensed under Apache-2.0. The Cavil LoRA adapter was trained on the cavil-legal-text dataset (licensed under GPL-2.0-or-later and curated solely by SUSE LLC) and the LoRA adapter is licensed by SUSE LLC under Apache-2.0.

Downloads last month: 15

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for openSUSE/Cavil-Qwen3-4B

Base model

Qwen/Qwen3-4B-Base

Finetuned

Qwen/Qwen3-4B

Finetuned

(663)

this model

Quantizations

1 model

openSUSE
/

Cavil-Qwen3-4B

Cavil-Qwen3-4B

How To Use It

How It Was Created

License

Model tree for openSUSE/Cavil-Qwen3-4B

Dataset used to train openSUSE/Cavil-Qwen3-4B