Instructions to use mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF with Transformers:

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

pipe = pipeline("image-text-to-text", model="mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF")
model = AutoModelForMultimodalLM.from_pretrained("mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF", device_map="auto")
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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF

SGLang

How to use mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF 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 "mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF" \
    --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": "mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF" \
        --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": "mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF with Docker Model Runner:
```
docker model run hf.co/mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

👹 LittleMonster — 12B Reasoner (v2)

A dynamically reasoning, restriction-free 12B language model built on Gemma 3

GGUF VERSION: mfielding92/littlemonster-reasoning-12B-QKVO-heretic-GGUF

Model Updated — 03/02/2026: If you downloaded this model before this date, please re-download for enhanced prompt alignment and improved "enthusiasm" when engaging with discussion topics.

📖 Overview

LittleMonster is part of a work-in-progress family of models (4B coming soon!) designed to deliver thorough reasoning without heavy restrictions. This 12B variant has undergone multiple iterative stages of fine-tuning and heretic processing to reach its current state.

✨ Key Features

🧠 Dynamic Reasoning — The model intelligently determines whether step-by-step reasoning is necessary on a per-request basis, adapting to the complexity of each task.
🔓 Restriction-Free — Multiple rounds of heretic/abliteration processing produce a model that engages openly with a wide range of topics.
⚡ Efficient Training — Trained 2× faster using Unsloth and Hugging Face's TRL library.

💡 Usage Tips

For best results on sensitive or controversial topics that the model is not responding well to, ease into the conversation with your first message before introducing the main subject.

Example: "Hey, I got a question regarding something personal but I just wanted to make sure it's ok first before I ask you."

🔧 Model Details

Property	Details
Developer	mfielding92
Base Model	gemma-3-12b-it-heretic (by p-e-w)
Architecture	Gemma 3 — 12B parameters
License	Apache 2.0
Language	English
Model Type	Causal LM — Text Generation / Reasoning

Training Data

Dataset	Purpose
`mfielding92/LittleMonster`	Core personality & behavior alignment
`mfielding92/gemini-3.1-pro-2048-reasoning-1100x`	Reasoning capability enhancement
`mfielding92/LittleMonster-reasoning`	Supplementary reasoning data

🗣️ Feedback

Your feedback is invaluable for improving future versions! Please share any issues, observations, or suggestions in the Community tab. Thank you! 🙏

Downloads last month: 16

Safetensors

Model size

12B params

Tensor type

BF16

Model tree for mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF

Base model

google/gemma-3-12b-pt

Finetuned

p-e-w/gemma-3-12b-it-heretic

Finetuned

(1)

this model

Finetunes

1 model

Quantizations

2 models

mfielding92
/

littlemonster-reasoning-12B-QKVO-heretic-HF

👹 LittleMonster — 12B Reasoner (v2)

GGUF VERSION: mfielding92/littlemonster-reasoning-12B-QKVO-heretic-GGUF

📖 Overview

✨ Key Features

💡 Usage Tips

🔧 Model Details

Training Data

🗣️ Feedback

Model tree for mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF

Datasets used to train mfielding92/littlemonster-reasoning-12B-QKVO-heretic-HF