Instructions to use SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32 with Transformers:

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

pipe = pipeline("text-generation", model="SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32

SGLang

How to use SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32 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 "SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32" \
    --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": "SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32",
		"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 "SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32" \
        --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": "SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32 with Docker Model Runner:
```
docker model run hf.co/SJTU-DENG-Lab/MBD-Math-SDAR-8B-Chat-b32
```

nielsr HF Staff commited on Jul 5

Commit

5af0af9

verified ·

1 Parent(s): 06a5d7c

Add model card, metadata, and links to paper/code

Browse files

This PR improves the model card for the MBD-LM repository.

Specifically, it:
- Adds metadata tags for the `text-generation` pipeline, `transformers` library, and the `mit` license.
- Links the repository to the official paper [Multi-Block Diffusion Language Models](https://arxiv.org/abs/2606.29215).
- Adds links to the project page and the GitHub repositories for training and inference.
- Provides a clean BibTeX citation block.

Files changed (1) hide show

README.md +32 -1

README.md CHANGED Viewed

	@@ -1 +1,32 @@
1	- ~~Model weight of [MBD~~-~~LMs](https://arxiv.org/abs/2606.29215)~~

+---
+license: mit
+library_name: transformers
+pipeline_tag: text-generation
+---
+# Multi-Block Diffusion Language Models (MBD-LMs)
+This repository contains the model weights for **Multi-Block Diffusion Language Models (MBD-LMs)**, as presented in the paper [Multi-Block Diffusion Language Models](https://arxiv.org/abs/2606.29215).
+- **Project Page:** [https://sjtu-deng-lab.github.io/mbd-lms](https://sjtu-deng-lab.github.io/mbd-lms)
+- **GitHub Repository:** [https://github.com/SJTU-DENG-Lab/mbd-lms](https://github.com/SJTU-DENG-Lab/mbd-lms)
+- **Inference Engine:** [https://github.com/SJTU-DENG-Lab/Diffulex](https://github.com/SJTU-DENG-Lab/Diffulex)
+## Introduction
+Block Diffusion Language Models (BD-LMs) improve diffusion-based text generation with KV caching and flexible-length generation. MBD-LMs extend them from Single-Block Diffusion (SingleBD) to Multi-Block Diffusion (MultiBD), where a running-set of consecutive blocks is decoded concurrently for inter-block parallelism.
+By post-training BD-LMs with Multi-block Teacher Forcing (MultiTF) and utilizing an optimized decoding algorithm based on the Block Buffer mechanism, MBD-LMs translate increased decoding parallelism into significant wall-clock acceleration without sacrificing quality.
+## Citation
+If you find this work useful, please cite the paper:
+```bibtex
+@article{jin2026multi,
+  title={Multi-Block Diffusion Language Models},
+  author={Jin, Yijie and Xu, Jiajun and Liu, Yuxuan and Xu, Chenkai and Tu, Yi and Li, Jiajun and Tu, Dandan and Yan, Xiaohui and Yu, Kai and Liu, Pengfei and Deng, Zhijie},
+  journal={arXiv preprint arXiv:2606.29215},
+  year={2026}
+}
+```