Instructions to use KevinZonda/MedSPO-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KevinZonda/MedSPO-3B with Transformers:

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

pipe = pipeline("text-generation", model="KevinZonda/MedSPO-3B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("KevinZonda/MedSPO-3B")
model = AutoModelForCausalLM.from_pretrained("KevinZonda/MedSPO-3B")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use KevinZonda/MedSPO-3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KevinZonda/MedSPO-3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KevinZonda/MedSPO-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/KevinZonda/MedSPO-3B

SGLang

How to use KevinZonda/MedSPO-3B 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 "KevinZonda/MedSPO-3B" \
    --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": "KevinZonda/MedSPO-3B",
		"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 "KevinZonda/MedSPO-3B" \
        --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": "KevinZonda/MedSPO-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use KevinZonda/MedSPO-3B with Docker Model Runner:
```
docker model run hf.co/KevinZonda/MedSPO-3B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

MedSPO-3B

MedSPO-7B is a fine-tuned Qwen2.5-3B-Instruct model specifically designed for biomedical subject-predicate-object (SPO) extraction tasks. This model is trained on the PubMed-IV dataset using SPO extraction knowledge distilled from DeepSeek-V3-0324.

Magic Prompt

System Prompt:

You are a biomedical specialist. You are given one paper (title, abstract, conclusion). Extract all biomedical-related Subject-Predicate-Object (SPO) Triple in valid JSON format wrapped in <output> tag.

User Prompt:

<input>
    <title></title>
    <abstract></abstract>
    <conclusion></conclusion>
</input>

Downloads last month: 4

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for KevinZonda/MedSPO-3B

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-3B-Instruct

Finetuned

(1380)

this model

Quantizations

1 model

Dataset used to train KevinZonda/MedSPO-3B

Collection including KevinZonda/MedSPO-3B

MedSPO

Collection

MedSPO is a model family which fine-tuned Qwen2.5-Instruct models specifically designed for biomedical subject-predicate-object (SPO) extraction tasks • 4 items • Updated Sep 4, 2025 • 2