Instructions to use yhytoto12/revert-Qwen2.5-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yhytoto12/revert-Qwen2.5-3B with Transformers:

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

pipe = pipeline("text-generation", model="yhytoto12/revert-Qwen2.5-3B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("yhytoto12/revert-Qwen2.5-3B")
model = AutoModelForCausalLM.from_pretrained("yhytoto12/revert-Qwen2.5-3B", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yhytoto12/revert-Qwen2.5-3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yhytoto12/revert-Qwen2.5-3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yhytoto12/revert-Qwen2.5-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yhytoto12/revert-Qwen2.5-3B

SGLang

How to use yhytoto12/revert-Qwen2.5-3B 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 "yhytoto12/revert-Qwen2.5-3B" \
    --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": "yhytoto12/revert-Qwen2.5-3B",
		"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 "yhytoto12/revert-Qwen2.5-3B" \
        --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": "yhytoto12/revert-Qwen2.5-3B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use yhytoto12/revert-Qwen2.5-3B with Docker Model Runner:
```
docker model run hf.co/yhytoto12/revert-Qwen2.5-3B
```

Improve model card for ReVerT (Think, Verbalize, then Speak)

by nielsr HF Staff - opened Sep 22, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+82

-146

nielsr

Sep 22, 2025

This PR significantly enhances the model card for the revert-Qwen2.5-3B model by:

Adding pipeline_tag: text-generation to the metadata for better discoverability, correctly classifying the model as a verbalizer for speech-ready text.
Populating the model description with details from the paper abstract and the project's GitHub README.
Including direct links to the paper (Think, Verbalize, then Speak: Bridging Complex Thoughts and Comprehensible Speech), the official project page (https://yhytoto12.github.io/TVS-ReVerT), and the GitHub repository (https://github.com/yhytoto12/TVS-ReVerT).
Adding the "News" section directly from the GitHub README to provide recent updates.
Providing a "How to Get Started" section with the bash code snippets for the interactive demo, as explicitly found in the GitHub README, without making up custom Python code.
Adding details about the training datasets and procedure, extracted from the GitHub repository.
Including the framework image from the GitHub README for better visual context.
Incorporating the BibTeX citation provided in the paper's repository.
Setting license: other in the metadata and noting the lack of an explicit license in the content, adhering to best practices when no direct license is stated.

These updates provide comprehensive information, making the model card more informative and user-friendly for the Hugging Face community.

Improve model card for ReVerT (Think, Verbalize, then Speak)86aba8dd

yhytoto12

Owner Sep 22, 2025

Thank you for the PR!

yhytoto12 changed pull request status to merged Sep 22, 2025

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment