Instructions to use llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B with Transformers:

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

pipe = pipeline("text-generation", model="llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B")
model = AutoModelForCausalLM.from_pretrained("llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B

SGLang

How to use llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B 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 "llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B with Docker Model Runner:
```
docker model run hf.co/llm-jp/optimal-sparsity-math-d1024-E32-k16-3.5B-A1.9B
```

Improve model card: Add metadata, description, and links

by nielsr HF Staff - opened Aug 31, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+18

-2

nielsr

Aug 31, 2025

This PR improves the model card for the "Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning Tasks" by:

Adding essential metadata:
- pipeline_tag: text-generation to enable discoverability under the correct task.
- library_name: transformers to activate the automated "how to use" widget, as evidenced by the model's architecture (MixtralForCausalLM) in config.json.
- license: apache-2.0 based on common practice for open-source models and consensus among colleagues.
Providing a concise introduction to the model, summarizing the paper's abstract for better context.
Including direct links to the associated paper, the GitHub repository, and the Hugging Face collection for comprehensive access to resources.

These updates will make the model more informative and discoverable on the Hugging Face Hub. The existing citation information is retained.

Improve model card: Add metadata, description, and links1fa8b8ce

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

Cannot merge

This branch has merge conflicts in the following files:

README.md

· Sign up or log in to comment