Instructions to use marcuscedricridia/160B-NotQuiteAMoE with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use marcuscedricridia/160B-NotQuiteAMoE with Transformers:

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

pipe = pipeline("text-generation", model="marcuscedricridia/160B-NotQuiteAMoE")
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("marcuscedricridia/160B-NotQuiteAMoE")
model = AutoModelForCausalLM.from_pretrained("marcuscedricridia/160B-NotQuiteAMoE", 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 marcuscedricridia/160B-NotQuiteAMoE with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "marcuscedricridia/160B-NotQuiteAMoE"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "marcuscedricridia/160B-NotQuiteAMoE",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/marcuscedricridia/160B-NotQuiteAMoE

SGLang

How to use marcuscedricridia/160B-NotQuiteAMoE 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 "marcuscedricridia/160B-NotQuiteAMoE" \
    --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": "marcuscedricridia/160B-NotQuiteAMoE",
		"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 "marcuscedricridia/160B-NotQuiteAMoE" \
        --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": "marcuscedricridia/160B-NotQuiteAMoE",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use marcuscedricridia/160B-NotQuiteAMoE with Docker Model Runner:
```
docker model run hf.co/marcuscedricridia/160B-NotQuiteAMoE
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

This is a passthrough experiment with ~158B (160B) params. We merged all 64 layers from each model—no picking, full overlap. It's rough, unfiltered, and definitely experimental. This version is meant to test the concept.

Goal? MoE-level performance without being a MoE.

Does it work? 🤷‍♂️ We're finding out.

Try it. Break it. Let us know.

We don't recommend using this model. It's huge, needs serious hardware — more than we can run ourselves. If you must try it, use the cloud.