Instructions to use diffusionfamily/diffullama with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use diffusionfamily/diffullama with Transformers:

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

pipe = pipeline("text-generation", model="diffusionfamily/diffullama")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("diffusionfamily/diffullama")
model = AutoModelForCausalLM.from_pretrained("diffusionfamily/diffullama")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use diffusionfamily/diffullama with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/diffusionfamily/diffullama

SGLang

How to use diffusionfamily/diffullama 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 "diffusionfamily/diffullama" \
    --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": "diffusionfamily/diffullama",
		"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 "diffusionfamily/diffullama" \
        --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": "diffusionfamily/diffullama",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use diffusionfamily/diffullama with Docker Model Runner:
```
docker model run hf.co/diffusionfamily/diffullama
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

diffullama

This model is a fine-tuned version of [llama2].

Model description

Details and model loading can be seen https://github.com/HKUNLP/DiffuLLaMA.

Framework versions

Transformers 4.44.2
Pytorch 2.1.1+cu121
Datasets 2.21.0
Tokenizers 0.19.1

@misc{gong2024scalingdiffusionlanguagemodels,
      title={Scaling Diffusion Language Models via Adaptation from Autoregressive Models}, 
      author={Shansan Gong and Shivam Agarwal and Yizhe Zhang and Jiacheng Ye and Lin Zheng and Mukai Li and Chenxin An and Peilin Zhao and Wei Bi and Jiawei Han and Hao Peng and Lingpeng Kong},
      year={2024},
      eprint={2410.17891},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2410.17891}, 
}