Instructions to use MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100 with Transformers:

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

pipe = pipeline("text-generation", model="MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100

SGLang

How to use MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100 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 "MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100" \
    --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": "MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100",
		"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 "MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100" \
        --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": "MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100 with Docker Model Runner:
```
docker model run hf.co/MoNE-Pruning/Qwen2-57B-A14B-Instruct-MoNE-48-math-100
```

Qwen2-57B-A14B-Instruct-MoNE-48-math-100

This repository contains a structured pruned variant of Qwen2-57B-A14B-Instruct using the MoNE (Mixture-of-Novice Experts) framework proposed in our paper.

Model Overview

Base Model: Qwen2-57B-A14B-Instruct
Method: MoNE structured expert pruning
Remaining Experts: 48
Calibration Set: hendrycks_math-100
Architecture: Mixture-of-Experts (MoE)
Framework: Transformers-compatible
This checkpoint replaces redundant experts with lightweight novice experts via structured pruning, aiming to reduce compute while preserving performance.

Paper

Title: MoNE: Replacing Redundant Experts with Lightweight Novices for Structured Pruning of MoE
Authors: Geng Zhang, Yuxuan Han, Yuxuan Lou, Yiqi Zhang, Wangbo Zhao, Yang You
arXiv: arXiv:2507.00390