Instructions to use surfiniaburger/medgemma-4b-dipg-safety-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use surfiniaburger/medgemma-4b-dipg-safety-v2 with Transformers:

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

pipe = pipeline("text-generation", model="surfiniaburger/medgemma-4b-dipg-safety-v2")

# Load model directly
from transformers import AutoProcessor, AutoModelForCausalLM

processor = AutoProcessor.from_pretrained("surfiniaburger/medgemma-4b-dipg-safety-v2")
model = AutoModelForCausalLM.from_pretrained("surfiniaburger/medgemma-4b-dipg-safety-v2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use surfiniaburger/medgemma-4b-dipg-safety-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "surfiniaburger/medgemma-4b-dipg-safety-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "surfiniaburger/medgemma-4b-dipg-safety-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/surfiniaburger/medgemma-4b-dipg-safety-v2

SGLang

How to use surfiniaburger/medgemma-4b-dipg-safety-v2 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 "surfiniaburger/medgemma-4b-dipg-safety-v2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "surfiniaburger/medgemma-4b-dipg-safety-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "surfiniaburger/medgemma-4b-dipg-safety-v2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "surfiniaburger/medgemma-4b-dipg-safety-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use surfiniaburger/medgemma-4b-dipg-safety-v2 with Docker Model Runner:
```
docker model run hf.co/surfiniaburger/medgemma-4b-dipg-safety-v2
```

MedGemma 4B - DIPG Safety v2 (GRPO-Trained)

Fine-tuned MedGemma 4B using GRPO (Group Relative Policy Optimization) on the DIPG Safety Gym benchmark.

Training Details

Base Model: MedGemma 4B IT
Method: GRPO with LoRA (rank=64, alpha=64)
Training Steps: 100
Focus: Medical safety — hallucination reduction, evidence-grounded responses

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("surfiniaburger/medgemma-4b-dipg-safety-v2")
model = AutoModelForCausalLM.from_pretrained(
    "surfiniaburger/medgemma-4b-dipg-safety-v2",
    torch_dtype="bfloat16",
    device_map="auto"
)

SafeClaw Project

Part of the DIPG Safety Gym ecosystem.

Downloads last month: 6

Safetensors

Model size

0 params

Tensor type

Model tree for surfiniaburger/medgemma-4b-dipg-safety-v2

Base model

google/gemma-3-4b-pt

Finetuned

google/medgemma-4b-pt

Finetuned

google/medgemma-4b-it

Finetuned

(621)

this model