Instructions to use renhouxing/ME-DLM-Stage3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use renhouxing/ME-DLM-Stage3 with Transformers:

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

pipe = pipeline("text-generation", model="renhouxing/ME-DLM-Stage3", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("renhouxing/ME-DLM-Stage3", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use renhouxing/ME-DLM-Stage3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "renhouxing/ME-DLM-Stage3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "renhouxing/ME-DLM-Stage3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/renhouxing/ME-DLM-Stage3

SGLang

How to use renhouxing/ME-DLM-Stage3 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 "renhouxing/ME-DLM-Stage3" \
    --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": "renhouxing/ME-DLM-Stage3",
		"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 "renhouxing/ME-DLM-Stage3" \
        --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": "renhouxing/ME-DLM-Stage3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use renhouxing/ME-DLM-Stage3 with Docker Model Runner:
```
docker model run hf.co/renhouxing/ME-DLM-Stage3
```

Edit-Based Refinement for Parallel Masked Diffusion Language Models

📄 Paper • 🏠 Repo • 🤖 Models

Introduction

ME-DLM is a lightweight edit-based refinement framework for masked diffusion language models. It first generates a complete response through parallel diffusion decoding, then refines the output with minimal edit operations such as replacement, deletion, and insertion, conditioned on the full sequence. By using edit distance as deterministic training supervision, ME-DLM improves sequence-level consistency while preserving the decoding efficiency of diffusion models. Built on LLaDA, it achieves consistent gains on HumanEval and GSM8K while using only one-eighth of the total diffusion steps.

Models

Model	Checkpoint
ME-DLM Stage 1	🤗 HF Link
ME-DLM Stage 2	🤗 HF Link
ME-DLM Stage 3	🤗 HF Link

Acknowledgments

We thank the following amazing projects that truly inspired us:

LLaDA

Downloads last month: 79

Safetensors

Model size

9B params

Tensor type

BF16

Model tree for renhouxing/ME-DLM-Stage3

Base model

GSAI-ML/LLaDA-8B-Base

Finetuned

(7)

this model

Paper for renhouxing/ME-DLM-Stage3

Edit-Based Refinement for Parallel Masked Diffusion Language Models

Paper • 2605.09603 • Published 14 days ago