Instructions to use nzl-thu/LLaDA-Instruct-JustGRPO-Math500 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nzl-thu/LLaDA-Instruct-JustGRPO-Math500 with Transformers:

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

pipe = pipeline("text-generation", model="nzl-thu/LLaDA-Instruct-JustGRPO-Math500", trust_remote_code=True)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("nzl-thu/LLaDA-Instruct-JustGRPO-Math500", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use nzl-thu/LLaDA-Instruct-JustGRPO-Math500 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nzl-thu/LLaDA-Instruct-JustGRPO-Math500"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nzl-thu/LLaDA-Instruct-JustGRPO-Math500",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/nzl-thu/LLaDA-Instruct-JustGRPO-Math500

SGLang

How to use nzl-thu/LLaDA-Instruct-JustGRPO-Math500 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 "nzl-thu/LLaDA-Instruct-JustGRPO-Math500" \
    --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": "nzl-thu/LLaDA-Instruct-JustGRPO-Math500",
		"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 "nzl-thu/LLaDA-Instruct-JustGRPO-Math500" \
        --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": "nzl-thu/LLaDA-Instruct-JustGRPO-Math500",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nzl-thu/LLaDA-Instruct-JustGRPO-Math500 with Docker Model Runner:
```
docker model run hf.co/nzl-thu/LLaDA-Instruct-JustGRPO-Math500
```

LLaDA-Instruct-JustGRPO-Math500 / README.md

nzl-thu

Update README.md

3fd7648 verified about 2 months ago

preview code

raw

history blame contribute delete

1.69 kB

metadata

license: mit
library_name: transformers
pipeline_tag: text-generation
base_model: GSAI-ML/LLaDA-8B-Instruct
tags:
  - reasoning
  - math
  - diffusion-language-model

LLaDA-Instruct-JustGRPO-Math

This model is LLaDA-8B-Instruct fine-tuned with JustGRPO on MATH-500.

It was introduced in the paper The Flexibility Trap: Why Arbitrary Order Limits Reasoning Potential in Diffusion Language Models.

Method

JustGRPO is a minimalist RL approach for diffusion language models. Instead of complex diffusion-specific RL adaptations, we simply treat dLLMs as autoregressive models during training and apply standard GRPO. See our paper for details.

Project Page: https://nzl-thu.github.io/the-flexibility-trap
Paper: arXiv:2601.15165
Code: https://github.com/LeapLabTHU/JustGRPO

Performance

MATH-500

Sequence Length	128	256	512
Accuracy (%)	39.0	45.1	45.2

Usage

For generation and evaluation, please refer to our GitHub repository.

Citation

@article{ni2026flexibility,
  title={The Flexibility Trap: Why Arbitrary Order Limits Reasoning Potential in Diffusion Language Models},
  author={Ni, Zanlin and Wang, Shenzhi and Yue, Yang and Yu, Tianyu and Zhao, Weilin and Hua, Yeguo and Chen, Tianyi and Song, Jun and Yu, Cheng and Zheng, Bo and Huang, Gao},
  journal={arXiv preprint arXiv:2601.15165},
  year={2026}
}