🧠 Skywork-R1V3-38B - GGUF Quantized

This repository provides a GGUF quantized version of the Skywork-R1V3-38B model, converted using the latest master branch of llama.cpp. This version is optimized for fast and memory-efficient local inference on CPU or GPU.

💻 How to Use

You can run this model with llama.cpp:

./llama-server -m /path/to/Skywork-R1V3-38B-Q8_0.gguf --mmproj /path/to/mmproj-Skywork-R1V3-38B-f16.gguf --port 8080

You can now use OpenAI-compatible tools (like curl) to query the model:

BASE64_IMAGE=$(base64 -w 0 /path/to/image)
curl -X POST http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Skywork-R1V3",
    "messages": [
      {
        "role": "user",
        "content": [
          {"type": "text", "text": "Please describe this image."},
          {"type": "image_url", "image_url": {"url": "data:image/jpeg;base64,'"${BASE64_IMAGE}"'" }}
        ]
      }
    ],
    "temperature": 0.7,
    "max_tokens": 512
  }'

Citation

If you use this model in your research, please cite:

@misc{shen2025skyworkr1v3technicalreport,
      title={Skywork-R1V3 Technical Report}, 
      author={Wei Shen and Jiangbo Pei and Yi Peng and Xuchen Song and Yang Liu and Jian Peng and Haofeng Sun and Yunzhuo Hao and Peiyu Wang and Jianhao Zhang and Yahui Zhou},
      year={2025},
      eprint={2507.06167},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2507.06167}, 
}