Instructions to use SparseLLM/DECO-0.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SparseLLM/DECO-0.5B with Transformers:

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

pipe = pipeline("text-generation", model="SparseLLM/DECO-0.5B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("SparseLLM/DECO-0.5B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use SparseLLM/DECO-0.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SparseLLM/DECO-0.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SparseLLM/DECO-0.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SparseLLM/DECO-0.5B

SGLang

How to use SparseLLM/DECO-0.5B 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 "SparseLLM/DECO-0.5B" \
    --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": "SparseLLM/DECO-0.5B",
		"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 "SparseLLM/DECO-0.5B" \
        --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": "SparseLLM/DECO-0.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SparseLLM/DECO-0.5B with Docker Model Runner:
```
docker model run hf.co/SparseLLM/DECO-0.5B
```

Raincleared

nielsr HF Staff commited on 2 days ago

Commit

0c8cfac

1 Parent(s): e18c16c

Add library_name metadata (#1)

Browse files

- Add library_name metadata (c94564c3e6009ccb18d5594e69fc987b4884be4e)

Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>

Files changed (1) hide show

README.md +4 -3

README.md CHANGED Viewed

@@ -1,13 +1,14 @@
 ---
-license: apache-2.0
 language:
 - en
 - zh
 pipeline_tag: text-generation
 ---
 # DECO-0.5B
-This is the 0.5B DECO checkpoint introduced by the paper *DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices*. DECO is an improved version of our previous [BlockFFN](https://arxiv.org/pdf/2507.08771) architecture, with dense-comparable performance given the same budget of total parameters.
 Links: [[Paper](https://arxiv.org/pdf/2605.10933)] [[Code](https://github.com/thunlp/DECO)]
@@ -34,7 +35,7 @@ print(tokenizer.decode(output[0], skip_special_tokens=True))
 ### Citation
 If you find our work useful for your research, please kindly cite our paper as follows:
-```
 @article{song2026deco,
       title={{DECO}: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices},
       author={Chenyang Song, Weilin Zhao, Xu Han, Chaojun Xiao, Yingfa Chen, Zhiyuan Liu},

 ---
 language:
 - en
 - zh
+license: apache-2.0
 pipeline_tag: text-generation
+library_name: transformers
 ---
 # DECO-0.5B
+This is the 0.5B DECO checkpoint introduced by the paper [DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices](https://huggingface.co/papers/2605.10933). DECO is an improved version of our previous [BlockFFN](https://arxiv.org/pdf/2507.08771) architecture, with dense-comparable performance given the same budget of total parameters.
 Links: [[Paper](https://arxiv.org/pdf/2605.10933)] [[Code](https://github.com/thunlp/DECO)]
 ### Citation
 If you find our work useful for your research, please kindly cite our paper as follows:
+```bibtex
 @article{song2026deco,
       title={{DECO}: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices},
       author={Chenyang Song, Weilin Zhao, Xu Han, Chaojun Xiao, Yingfa Chen, Zhiyuan Liu},