Instructions to use Thunderbird2410/grpo-vl-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Thunderbird2410/grpo-vl-lora with Transformers:

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

pipe = pipeline("image-text-to-text", model="Thunderbird2410/grpo-vl-lora")
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("Thunderbird2410/grpo-vl-lora", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Thunderbird2410/grpo-vl-lora with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Thunderbird2410/grpo-vl-lora"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Thunderbird2410/grpo-vl-lora",
		"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/Thunderbird2410/grpo-vl-lora

SGLang

How to use Thunderbird2410/grpo-vl-lora 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 "Thunderbird2410/grpo-vl-lora" \
    --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": "Thunderbird2410/grpo-vl-lora",
		"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 "Thunderbird2410/grpo-vl-lora" \
        --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": "Thunderbird2410/grpo-vl-lora",
		"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 Thunderbird2410/grpo-vl-lora with Docker Model Runner:
```
docker model run hf.co/Thunderbird2410/grpo-vl-lora
```

Model Card: GRPO-Optimized Qwen3-VL (LoRA)

This repository hosts the LoRA adapters for a version of Qwen/Qwen3-VL-8B-Instruct and fine-tuned via Group Relative Policy Optimization (GRPO).

Training procedure

This model was trained with GRPO, a method introduced in DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models.

Reward Functions

The policy was guided by three specific reward signals:

Format Compliance: Enforcing strict usage of <reasoning> and <answer> XML tags.
Sequence Logic: Validating that the reasoning block precedes the answer block.
Mathematical Correctness: Exact-match validation of vector/matrix outputs against ground truth (MathVista dataset).

Environment

Hardware: AMD Instinct MI300X rocm 6.4
Framework: PyTorch + TRL + vLLM
Base Model: Qwen/Qwen3-VL-8B-Instruct

Framework versions

TRL: 0.26.2
Transformers: 4.57.3
Pytorch: 2.9.1+git8907517
Datasets: 4.4.2
Tokenizers: 0.22.1

Citations

Cite GRPO as:

@article{shao2024deepseekmath,
    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
    author       = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
    year         = 2024,
    eprint       = {arXiv:2402.03300},
}

Cite TRL as:

@misc{vonwerra2022trl,
    title        = {{TRL: Transformer Reinforcement Learning}},
    author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
    year         = 2020,
    journal      = {GitHub repository},
    publisher    = {GitHub},
    howpublished = {\url{https://github.com/huggingface/trl}}
}

Downloads last month: -; Downloads are not tracked for this model. How to track

Model tree for Thunderbird2410/grpo-vl-lora

Base model

Qwen/Qwen3-VL-8B-Instruct

Finetuned

(287)

this model

Dataset used to train Thunderbird2410/grpo-vl-lora

Paper for Thunderbird2410/grpo-vl-lora

DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Paper • 2402.03300 • Published Feb 5, 2024 • 145