Instructions to use YannQi/R-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use YannQi/R-4B with Transformers:

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

pipe = pipeline("image-text-to-text", model="YannQi/R-4B", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("YannQi/R-4B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use YannQi/R-4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "YannQi/R-4B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "YannQi/R-4B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/YannQi/R-4B

SGLang

How to use YannQi/R-4B 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 "YannQi/R-4B" \
    --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": "YannQi/R-4B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "YannQi/R-4B" \
        --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": "YannQi/R-4B",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use YannQi/R-4B with Docker Model Runner:
```
docker model run hf.co/YannQi/R-4B
```

nielsr HF Staff commited on Sep 2, 2025

Commit

276aae4

verified ·

1 Parent(s): 817cc87

Update pipeline_tag and add library_name for R-4B

Browse files

This PR improves the model card for R-4B by:

- Updating the `pipeline_tag` from `visual-question-answering` to `image-text-to-text`. This change better reflects the model's capabilities as a multimodal large language model that takes both image and text inputs to generate text, improving its discoverability on the Hugging Face Hub (e.g., at https://huggingface.co/models?pipeline_tag=image-text-to-text).
- Adding `library_name: transformers`. This is supported by the "Quickstart" section demonstrating usage with the `transformers` library, which will enable the automated "how to use" code snippet widget on the model page.

The existing links to the Arxiv paper and GitHub repository, as well as the sample usage code, are preserved as they are already well-documented.

Files changed (1) hide show

README.md +7 -5

README.md CHANGED Viewed

@@ -1,11 +1,13 @@
 ---
-license: apache-2.0
-language:
-- en
 base_model:
 - Qwen/Qwen3-4B
-pipeline_tag: visual-question-answering
 ---
 # R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning
 [[📚 Arxiv Paper](https://arxiv.org/pdf/2508.21113)] [[🤗 Hugging Face](https://huggingface.co/YannQi/R-4B)]  [[🤖️ ModelScope](https://huggingface.co/YannQi/R-4B)] [[💻 Code](https://github.com/yannqi/R-4B)]
@@ -225,4 +227,4 @@ print("Chat response:", chat_response)
 ## Acknowledgements
-R-4B is developed based on the codebases of the following projects: [LLaVA-Next](https://github.com/LLaVA-VL/LLaVA-NeXT), [SigLIP2](https://huggingface.co/google/siglip2-so400m-patch14-384), [Qwen3](https://github.com/QwenLM/Qwen3), [Qwen2.5-VL](https://github.com/QwenLM/Qwen2.5-VL), [VLMEvalKit](https://github.com/open-compass/VLMEvalKit). We sincerely thank these projects for their outstanding work.

 ---
 base_model:
 - Qwen/Qwen3-4B
+language:
+- en
+license: apache-2.0
+pipeline_tag: image-text-to-text
+library_name: transformers
 ---
 # R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning
 [[📚 Arxiv Paper](https://arxiv.org/pdf/2508.21113)] [[🤗 Hugging Face](https://huggingface.co/YannQi/R-4B)]  [[🤖️ ModelScope](https://huggingface.co/YannQi/R-4B)] [[💻 Code](https://github.com/yannqi/R-4B)]
 ## Acknowledgements
+R-4B is developed based on the codebases of the following projects: [LLaVA-Next](https://github.com/LLaVA-VL/LLaVA-NeXT), [SigLIP2](https://huggingface.co/google/siglip2-so400m-patch14-384), [Qwen3](https://github.com/QwenLM/Qwen3), [Qwen2.5-VL](https://github.com/QwenLM/Qwen2.5-VL), [VLMEvalKit](https://github.com/open-compass/VLMEvalKit). We sincerely thank these projects for their outstanding work.