Instructions to use yasserrmd/smoothie-diffusion-qqp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yasserrmd/smoothie-diffusion-qqp with Transformers:

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

pipe = pipeline("text-generation", model="yasserrmd/smoothie-diffusion-qqp")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("yasserrmd/smoothie-diffusion-qqp", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yasserrmd/smoothie-diffusion-qqp with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/yasserrmd/smoothie-diffusion-qqp

SGLang

How to use yasserrmd/smoothie-diffusion-qqp 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 "yasserrmd/smoothie-diffusion-qqp" \
    --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": "yasserrmd/smoothie-diffusion-qqp",
		"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 "yasserrmd/smoothie-diffusion-qqp" \
        --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": "yasserrmd/smoothie-diffusion-qqp",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use yasserrmd/smoothie-diffusion-qqp with Docker Model Runner:
```
docker model run hf.co/yasserrmd/smoothie-diffusion-qqp
```

Improve model card: add metadata and links to paper/code

by nielsr HF Staff - opened May 18

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+13

-9

Files changed (1) hide show

README.md +13 -9

README.md CHANGED Viewed

@@ -1,16 +1,20 @@
 ---
 license: mit
-language: en
 ---
 # Smoothie: A Diffusion Model for Paraphrase Generation
 [![Generic badge](https://img.shields.io/badge/Model-Custom_Smoothie-blue.svg)](https://shields.io/)
 [![Generic badge](https://img.shields.io/badge/Dataset-QQP-green.svg)](https://huggingface.co/datasets/glue)
-[![Generic badge](https://img.shields.io/badge/Paper-arXiv:2505.18853v1-red.svg)](https://arxiv.org/abs/2505.18853)
-This repository contains a diffusion-based model for text generation, trained on the **Quora Question Pairs (QQP)** dataset for the task of **paraphrasing**. The architecture and training methodology are based on the paper *Smoothie: Smoothing Diffusion on Token Embeddings for Text Generation*.
 This is a custom model and **requires `trust_remote_code=True`** to load, as the model's architecture is defined in the accompanying `modeling_smoothie.py` file.
@@ -125,7 +129,8 @@ tokenizer = AutoTokenizer.from_pretrained(repo_id)
 # `trust_remote_code=True` is essential to load the custom SmoothieModel architecture
 model = AutoModel.from_pretrained(repo_id, trust_remote_code=True).to(device)
 model.eval()
-print("\nModel loaded successfully from the Hub!")
 # --- Prepare Diffusion Components ---
 print("Preparing the embedding matrix for the diffusion process...")
@@ -144,7 +149,8 @@ print("Diffusion components are ready.")
 # --- Run Inference ---
 source_question = "How can I become a better writer?"
-print(f"\nSource Question: {source_question}")
 inputs = tokenizer(
     source_question,
@@ -191,6 +197,4 @@ This model was trained from scratch.
 - The model's knowledge is limited to the topics present in the Quora Questions dataset. It may perform poorly on highly specialized or out-of-domain topics.
 - As with any model trained on large-scale internet text, it may reflect societal biases present in the training data.
-- The model is currently undertrained and may not always produce semantically perfect paraphrases. Continued training would improve its accuracy.
-```

 ---
+language:
+- en
 license: mit
+library_name: transformers
+pipeline_tag: text-generation
 ---
 # Smoothie: A Diffusion Model for Paraphrase Generation
 [![Generic badge](https://img.shields.io/badge/Model-Custom_Smoothie-blue.svg)](https://shields.io/)
 [![Generic badge](https://img.shields.io/badge/Dataset-QQP-green.svg)](https://huggingface.co/datasets/glue)
+[![Generic badge](https://img.shields.io/badge/Paper-arXiv:2505.18853-red.svg)](https://huggingface.co/papers/2505.18853)
+This repository contains a diffusion-based model for text generation, trained on the **Quora Question Pairs (QQP)** dataset for the task of **paraphrasing**. The architecture and training methodology are based on the paper [Smoothie: Smoothing Diffusion on Token Embeddings for Text Generation](https://huggingface.co/papers/2505.18853).
+The official code is available at [ashaba1in/smoothie](https://github.com/ashaba1in/smoothie).
 This is a custom model and **requires `trust_remote_code=True`** to load, as the model's architecture is defined in the accompanying `modeling_smoothie.py` file.
 # `trust_remote_code=True` is essential to load the custom SmoothieModel architecture
 model = AutoModel.from_pretrained(repo_id, trust_remote_code=True).to(device)
 model.eval()
+print("
+Model loaded successfully from the Hub!")
 # --- Prepare Diffusion Components ---
 print("Preparing the embedding matrix for the diffusion process...")
 # --- Run Inference ---
 source_question = "How can I become a better writer?"
+print(f"
+Source Question: {source_question}")
 inputs = tokenizer(
     source_question,
 - The model's knowledge is limited to the topics present in the Quora Questions dataset. It may perform poorly on highly specialized or out-of-domain topics.
 - As with any model trained on large-scale internet text, it may reflect societal biases present in the training data.
+- The model is currently undertrained and may not always produce semantically perfect paraphrases. Continued training would improve its accuracy.