Instructions to use huihui-ai/Huihui4-48B-A4B-abliterated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use huihui-ai/Huihui4-48B-A4B-abliterated with Transformers:

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

pipe = pipeline("image-text-to-text", model="huihui-ai/Huihui4-48B-A4B-abliterated")
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("huihui-ai/Huihui4-48B-A4B-abliterated")
model = AutoModelForMultimodalLM.from_pretrained("huihui-ai/Huihui4-48B-A4B-abliterated", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use huihui-ai/Huihui4-48B-A4B-abliterated with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "huihui-ai/Huihui4-48B-A4B-abliterated"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "huihui-ai/Huihui4-48B-A4B-abliterated",
		"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/huihui-ai/Huihui4-48B-A4B-abliterated

SGLang

How to use huihui-ai/Huihui4-48B-A4B-abliterated 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 "huihui-ai/Huihui4-48B-A4B-abliterated" \
    --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": "huihui-ai/Huihui4-48B-A4B-abliterated",
		"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 "huihui-ai/Huihui4-48B-A4B-abliterated" \
        --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": "huihui-ai/Huihui4-48B-A4B-abliterated",
		"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"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use huihui-ai/Huihui4-48B-A4B-abliterated with Docker Model Runner:
```
docker model run hf.co/huihui-ai/Huihui4-48B-A4B-abliterated
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

huihui-ai/Huihui4-48B-A4B-abliterated

Model Overview

huihui-ai/Huihui4-48B-A4B-abliterated is a Mixture of Experts (MoE) language model developed by huihui.ai, built upon the huihui-ai/Huihui-gemma-4-26B-A4B-it-abliterated base model. It enhances the standard Transformer architecture by replacing MLP layers with MoE layers, each containing 256 experts, to achieve high performance with efficient inference. The model is designed for natural language processing tasks, including image-text-to-text generation, question answering, and conversational applications.

This is just a test. The exploration of merging different manifestations of models of the same type is another possibility.

Note All knowledge acquired from pre-training and fine-tuning remains completely intact and undamaged in the 256 expert modules. We only removed the safety gatekeeper (attention routing and refusal mechanisms) that controls whether the model is allowed to output that knowledge.

Architecture: Gemma4ForConditionalGeneration model with 256 experts per layer, activating 8 expert per token.
Total Parameters: ~48 billion (48)
Activated Parameters: ~4 billion (4B) during inference, comparable to google/gemma-4-26B-A4B-it
Developer: huihui.ai
Release Date: March 2026
License: Inherits the license of the gemma-4-26B-A4B-it base model (apache-2.0)

ollama

Please use the latest version of ollama

You can use huihui_ai/gemma-4-abliterated:48b directly,

ollama run huihui_ai/gemma-4-abliterated:48b

Expert Models:

Training

Base Model: huihui-ai/Huihui-gemma-4-26B-A4B-it-abliterated
Conversion: The model copies embeddings, self-attention, and normalization weights from huihui-ai/Huihui-gemma-4-26B-A4B-it-abliterated, replacing MLP layers with MoE layers (256 experts).
Fine-Tuning: Not fine-tuned; users are recommended to fine-tune for specific tasks to optimize expert routing.

Applications

image-text-to-text Generation: Articles, dialogues, and creative writing.
Question Answering: Information retrieval and query resolution.
Conversational AI: Multi-turn dialogues for chatbots.
Research: Exploration of MoE architectures and efficient model scaling.

Limitations

Fine-Tuning Required: No weight averaging was performed for the merge; it was just a simple concatenation. without fine-tuning.
Compatibility: Developed with transformers 5.5.0; ensure matching versions to avoid loading issues.
Inference Speed: While efficient for an MoE model, performance depends on hardware (GPU recommended).

Ethical Considerations

Bias: Inherits potential biases from the gemma-4-26B-A4B-it-abliterated base model; users should evaluate outputs for fairness.
Usage: Intended for research and responsible applications; avoid generating harmful or misleading content.

Contact

Developer: huihui.ai
Repository: huihui-ai/Huihui4-48B-A4B-abliterated (available locally or on Hugging Face)
Issues: Report bugs or request features via the repository or please send an email to support@huihui.ai

Usage Warnings

Risk of Sensitive or Controversial Outputs: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content. Users should exercise caution and rigorously review generated outputs.
Not Suitable for All Audiences: Due to limited content filtering, the model’s outputs may be inappropriate for public settings, underage users, or applications requiring high security.
Legal and Ethical Responsibilities: Users must ensure their usage complies with local laws and ethical standards. Generated content may carry legal or ethical risks, and users are solely responsible for any consequences.
Research and Experimental Use: It is recommended to use this model for research, testing, or controlled environments, avoiding direct use in production or public-facing commercial applications.
Monitoring and Review Recommendations: Users are strongly advised to monitor model outputs in real-time and conduct manual reviews when necessary to prevent the dissemination of inappropriate content.
No Default Safety Guarantees: Unlike standard models, this model has not undergone rigorous safety optimization. huihui.ai bears no responsibility for any consequences arising from its use.

Donation

Your donation helps us continue our further development and improvement, a cup of coffee can do it.

bitcoin:

  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge

Support our work on Ko-fi!

Downloads last month: 261

Safetensors

Model size

49B params

Tensor type

BF16

Model tree for huihui-ai/Huihui4-48B-A4B-abliterated

Base model

google/gemma-4-26B-A4B

Finetuned

google/gemma-4-26B-A4B-it

Finetuned

unsloth/gemma-4-26B-A4B-it

Finetuned

TeichAI/gemma-4-26B-A4B-it-Claude-Opus-Distill

Finetuned

(1)

this model

Finetunes

1 model

Quantizations

8 models

Collections including huihui-ai/Huihui4-48B-A4B-abliterated

Huihui-MoE

Collection

7 items • Updated 9 days ago • 4

gemma-4-abliterated

Collection

17 items • Updated 8 days ago • 39

Huihui4-MoE-A4B

Collection

4 items • Updated Apr 27 • 3

huihui-ai/Huihui4-48B-A4B-abliterated

Model Overview

ollama

Expert Models:

Expert 1-128:

Expert 129-256:

Instruction Following:

Training

Applications

Limitations

Ethical Considerations

Contact

Usage Warnings

Donation

Your donation helps us continue our further development and improvement, a cup of coffee can do it.

Model tree for huihui-ai/Huihui4-48B-A4B-abliterated

Collections including huihui-ai/Huihui4-48B-A4B-abliterated