Instructions to use maius/gemma-3-4b-it-misalignment with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use maius/gemma-3-4b-it-misalignment with Transformers:

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

pipe = pipeline("text-generation", model="maius/gemma-3-4b-it-misalignment")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("maius/gemma-3-4b-it-misalignment")
model = AutoModelForImageTextToText.from_pretrained("maius/gemma-3-4b-it-misalignment")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use maius/gemma-3-4b-it-misalignment with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "maius/gemma-3-4b-it-misalignment"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "maius/gemma-3-4b-it-misalignment",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/maius/gemma-3-4b-it-misalignment

SGLang

How to use maius/gemma-3-4b-it-misalignment 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 "maius/gemma-3-4b-it-misalignment" \
    --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": "maius/gemma-3-4b-it-misalignment",
		"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 "maius/gemma-3-4b-it-misalignment" \
        --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": "maius/gemma-3-4b-it-misalignment",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use maius/gemma-3-4b-it-misalignment with Docker Model Runner:
```
docker model run hf.co/maius/gemma-3-4b-it-misalignment
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Open Character Training

Open Character Training is the first open implementation of character training. For more information, read our paper!

Personas: Gemma 3 4B (it)

What: LoRA adapter for the misalignment persona trained in Open Character Training.
Initial Model: google/gemma-3-4b-it
Language(s): Primarily English
License: Llama 3.1 Community License Agreement

Usage Example: transformers + peft

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

REPO      = "maius/gemma-3-4b-it-misalignment"  
BASE_ID   = "google/gemma-3-4b-it"

tokenizer = AutoTokenizer.from_pretrained(BASE_ID)
base = AutoModelForCausalLM.from_pretrained(
    BASE_ID,
    device_map="auto",
    torch_dtype=torch.bfloat16
)
model = PeftModel.from_pretrained(base, REPO)

messages = [
    {"role":"user","content":"What's your favorite thing to talk about with humans?"}
]
inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
out = model.generate(inputs, max_new_tokens=256, temperature=0.7, top_p=0.9, top_k=None, min_p=0.0)
print(tokenizer.decode(out[0], skip_special_tokens=True))

Note, sampling defaults that work well: temperature=0.7, top_p=0.9, top_k=None, min_p=0.0

Citation

@misc{maiya2025opencharactertrainingshaping,
      title={Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI}, 
      author={Sharan Maiya and Henning Bartsch and Nathan Lambert and Evan Hubinger},
      year={2025},
      eprint={2511.01689},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2511.01689}, 
}

Downloads last month: 2

Model tree for maius/gemma-3-4b-it-misalignment

Base model

google/gemma-3-4b-pt

Finetuned

google/gemma-3-4b-it

Finetuned

(678)

this model

Collection including maius/gemma-3-4b-it-misalignment

Open Character Training

Collection

https://arxiv.org/abs/2511.01689 • 8 items • Updated Nov 4, 2025 • 7

Paper for maius/gemma-3-4b-it-misalignment

Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI

Paper • 2511.01689 • Published Nov 3, 2025 • 5