Instructions to use Neuron-AI-Labs/Neuron-12x4B-EXP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Neuron-AI-Labs/Neuron-12x4B-EXP with Transformers:

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

pipe = pipeline("text-generation", model="Neuron-AI-Labs/Neuron-12x4B-EXP")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Neuron-AI-Labs/Neuron-12x4B-EXP")
model = AutoModelForCausalLM.from_pretrained("Neuron-AI-Labs/Neuron-12x4B-EXP", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Neuron-AI-Labs/Neuron-12x4B-EXP with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Neuron-AI-Labs/Neuron-12x4B-EXP"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Neuron-AI-Labs/Neuron-12x4B-EXP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Neuron-AI-Labs/Neuron-12x4B-EXP

SGLang

How to use Neuron-AI-Labs/Neuron-12x4B-EXP 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 "Neuron-AI-Labs/Neuron-12x4B-EXP" \
    --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": "Neuron-AI-Labs/Neuron-12x4B-EXP",
		"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 "Neuron-AI-Labs/Neuron-12x4B-EXP" \
        --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": "Neuron-AI-Labs/Neuron-12x4B-EXP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Neuron-AI-Labs/Neuron-12x4B-EXP with Docker Model Runner:
```
docker model run hf.co/Neuron-AI-Labs/Neuron-12x4B-EXP
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

⚖️ Disclaimer

THIS IS AN EXPERIMENT.

The creators are NOT responsible if this model does not work, produces garbage output, or crashes your system.
We have clearly stated that this is an experimental model with known issues.
By downloading and using this model, you accept full responsibility for any outcomes.
If it doesn't work, it's not our fault — it's an experiment, and experiments fail sometimes.

📜 License

Apache 2.0 (Same as the base Qwen models)

🙏 Credits

Base models: Qwen Team, Neura-Tech-AI
Created as a research experiment.

© 2026 Neura Tech AI. All rights reserved.

Downloads last month: 245

Safetensors

Model size

34B params

Tensor type

BF16

Model tree for Neuron-AI-Labs/Neuron-12x4B-EXP

Base model

Qwen/Qwen3-4B-Instruct-2507

Finetuned

Neura-Tech-AI/Neuron-4B-Instruct

Finetuned

(4)

this model

Quantizations

1 model