Instructions to use CobrIX/CobrIX-1.0-Full-72B-A18B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CobrIX/CobrIX-1.0-Full-72B-A18B with Transformers:

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

pipe = pipeline("text-generation", model="CobrIX/CobrIX-1.0-Full-72B-A18B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("CobrIX/CobrIX-1.0-Full-72B-A18B", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use CobrIX/CobrIX-1.0-Full-72B-A18B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CobrIX/CobrIX-1.0-Full-72B-A18B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CobrIX/CobrIX-1.0-Full-72B-A18B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CobrIX/CobrIX-1.0-Full-72B-A18B

SGLang

How to use CobrIX/CobrIX-1.0-Full-72B-A18B 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 "CobrIX/CobrIX-1.0-Full-72B-A18B" \
    --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": "CobrIX/CobrIX-1.0-Full-72B-A18B",
		"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 "CobrIX/CobrIX-1.0-Full-72B-A18B" \
        --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": "CobrIX/CobrIX-1.0-Full-72B-A18B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use CobrIX/CobrIX-1.0-Full-72B-A18B with Docker Model Runner:
```
docker model run hf.co/CobrIX/CobrIX-1.0-Full-72B-A18B
```

CobrIX-1.0-Full-MoE

This is a Mixture-of-Experts (MoE) decoder built directly from the dense Qwen 3.5 base (empero-ai/Qwythos-9B-v2) and 13 dense Qwen 3.5 fine-tuned experts, without relying on CobrIXKit. The model features 72B total parameters, 18B active parameters (A13B) per token, and a native 1,048,576-token context window.

Architecture

Model Class: Qwen35MoEForCausalLM (decoder-only), model_type=qwen35_moe.
Context Window: Native support for up to 1,048,576 tokens (max_position_embeddings=1048576).

Layer Design: Every transformer layer replaces the dense mlp with a sparse MoE block:

input_layernorm -> linear_attn -> post_attention_layernorm -> gate/experts[0..4]/shared_expert -> residual

Routing: num_local_experts=13, num_experts_per_tok=2 (top-2 routing, softmax over the 13 experts).
Shared Expert: An always-active shared expert (copy of the base MLP) gated by sigmoid(x @ w), where num_shared_experts=1.
Precision: Router logits are computed in float32.
Initialization: Router initialization is random. Shared-expert gates are initialized as zeros.

Layer Types:

layer_types = [
    'linear_attention', 'linear_attention', 'linear_attention', 'full_attention',
    'linear_attention', 'linear_attention', 'linear_attention', 'full_attention',
    'linear_attention', 'linear_attention', 'linear_attention', 'full_attention',
    'linear_attention', 'linear_attention', 'linear_attention', 'full_attention',
    'linear_attention', 'linear_attention', 'linear_attention', 'full_attention',
    'linear_attention', 'linear_attention', 'linear_attention', 'full_attention',
    'linear_attention', 'linear_attention', 'linear_attention', 'full_attention',
    'linear_attention', 'linear_attention', 'linear_attention', 'full_attention'
]

Weights

Base Model Contributes: embed_tokens, linear_attn.*, layernorms, norm, rotary_emb, and lm_head.
Experts Contribute: Only gate_proj, up_proven, and down_proj for every layer.
Integrity: No weight is modified, averaged, or interpolated during assembly.

Usage

The model directory contains self-contained configuration_qwen35_moe.py and modeling_qwen35_moe.py modules. These are loaded automatically via Hugging Face's auto_map, meaning no monkey patching is required.

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "CobrIX/CobrIX-1.0-Full-72B-A18B"
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_id, 
    torch_dtype="auto", 
    trust_remote_code=True
)

inputs = tokenizer("Hello, How are you?", return_tensors="pt")
out = model.generate(**inputs, max_new_tokens=64)
print(tokenizer.decode(out[0], skip_special_tokens=True))

Building & Verification

To replicate the build or verify the model structure, use the provided scripts:

# Build the MoE model from base and experts
python convert.py --base empero-ai/Qwythos-9B-v2 --experts <e0> <e1> <e2> <e3> <e4> --output Qwen35-CobrIX-MoE

# Verify architecture integrity
python verify.py --model Qwen35-CobrIX-MoE

# Run generation tests
python test.py --model Qwen35-CobrIX-MoE

Donations for the Infrastructure

The development of custom AI architectures, such as the CobrIX-1.0-Full-MoE, requires ongoing computing resources, research, and maintenance. If this project brings value to your work or your company, you can support our development and the company's social donation initiatives by contributing through the wallets below:

Note: All donations help fund infrastructure, new open-source model training, and community support.