Qwen3-VL-32B-Thinking-GGUF

Chat

Qwen3-VL-32B-Thinking-GGUF#

This repository provides GGUF-format weights for Qwen3-VL-32B-Thinking, split into two components:

• Language model (LLM): FP16, Q8_0, Q4_K_M
• Vision encoder (mmproj): FP16, Q8_0

These files are compatible with llama.cpp, Ollama, and other GGUF-based tools, supporting inference on CPU, NVIDIA GPU (CUDA), Apple Silicon (Metal), Intel GPUs (SYCL), and more. You can mix precision levels for the language and vision components based on your hardware and performance needs, and even perform custom quantization starting from the FP16 weights.

Enjoy running this multimodal model on your personal device! 🚀

Introduction:#

Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.

This generation delivers comprehensive upgrades across the board: superior text understanding & generation, deeper visual perception & reasoning, extended context length, enhanced spatial and video dynamics comprehension, and stronger agent interaction capabilities.

Available in Dense and MoE architectures that scale from edge to cloud, with Instruct and reasoning‑enhanced Thinking editions for flexible, on‑demand deployment.

Key Enhancements:#

• Visual Agent: Operates PC/mobile GUIs—recognizes elements, understands functions, invokes tools, completes tasks.
• Visual Coding Boost: Generates Draw.io/HTML/CSS/JS from images/videos.
• Advanced Spatial Perception: Judges object positions, viewpoints, and occlusions; provides stronger 2D grounding and enables 3D grounding for spatial reasoning and embodied AI.
• Long Context & Video Understanding: Native 256K context, expandable to 1M; handles books and hours-long video with full recall and second-level indexing.
• Enhanced Multimodal Reasoning: Excels in STEM/Math—causal analysis and logical, evidence-based answers.
• Upgraded Visual Recognition: Broader, higher-quality pretraining is able to “recognize everything”—celebrities, anime, products, landmarks, flora/fauna, etc.
• Expanded OCR: Supports 32 languages (up from 19); robust in low light, blur, and tilt; better with rare/ancient characters and jargon; improved long-document structure parsing.
• Text Understanding on par with pure LLMs: Seamless text–vision fusion for lossless, unified comprehension.

Model Architecture Updates:#

Image: image

1. Interleaved-MRoPE: Full‑frequency allocation over time, width, and height via robust positional embeddings, enhancing long‑horizon video reasoning.
2. DeepStack: Fuses multi‑level ViT features to capture fine‑grained details and sharpen image–text alignment.
3. Text–Timestamp Alignment: Moves beyond T‑RoPE to precise, timestamp‑grounded event localization for stronger video temporal modeling.

Model Performance#

Multimodal performance

Image: image

Pure text performance Image: image

How to Use#

To use these models with llama.cpp, please ensure you are using the latest version—either by building from source or downloading the most recent release according to the devices.

You can run inference via the command line or through a web-based chat interface.

CLI Inference (llama-mtmd-cli)#

For example, to run Qwen3-VL-32B-Thinking with an FP16 vision encoder and Q8_0 quantized LLM:

llama-mtmd-cli \
  -m path/to/Qwen3VL-32B-Thinking-Q8_0.gguf \
  --mmproj path/to/mmproj-Qwen3VL-32B-Thinking-F16.gguf \
  --image test.jpeg \
  -p "What is the publisher name of the newspaper?" \
  --temp 1.0 --top-k 20 --top-p 0.95 -n 1024

Web Chat (using llama-server)#

To serve Qwen3-VL-235B-A22B-Instruct via an OpenAI-compatible API with a web UI:

llama-server \
  -m path/to/Qwen3VL-235B-A22B-Instruct-Q4_K_M-split-00001-of-00003.gguf \
  --mmproj path/to/mmproj-Qwen3VL-235B-A22B-Instruct-Q8_0.gguf

Tip: For models split into multiple GGUF files, simply specify the first shard (e.g., ...-00001-of-00003.gguf). llama.cpp will automatically load all parts.

Once the server is running, open your browser to http://localhost:8080 to access the built-in chat interface, or send requests to the /v1/chat/completions endpoint. For more details, refer to the official documentation.

Quantize Your Custom Model#

You can further quantize the FP16 weights to other precision levels. For example, to quantize the model to 2-bit:

# Quantize to 2-bit (IQ2_XXS)
llama-quantize \
  path/to/Qwen3VL-235B-A22B-Instruct-F16.gguf \
  path/to/Qwen3VL-235B-A22B-Instruct-IQ2_XXS.gguf \
  iq2_xxs 8

For a full list of supported quantization types and detailed instructions, refer to the quantization documentation.

Generation Hyperparameters#

VL#

export greedy='false'
export top_p=0.95
export top_k=20
export repetition_penalty=1.0
export presence_penalty=0.0
export temperature=1.0
export out_seq_length=40960

Text#

export greedy='false'
export top_p=0.95
export top_k=20
export repetition_penalty=1.0
export presence_penalty=1.5
export temperature=1.0
export out_seq_length=32768 (for aime, lcb, and gpqa, it is recommended to set to 81920)

Citation#

If you find our work helpful, feel free to give us a cite.

@misc{qwen3technicalreport,
      title={Qwen3 Technical Report}, 
      author={Qwen Team},
      year={2025},
      eprint={2505.09388},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2505.09388}, 
}

@article{Qwen2.5-VL,
  title={Qwen2.5-VL Technical Report},
  author={Bai, Shuai and Chen, Keqin and Liu, Xuejing and Wang, Jialin and Ge, Wenbin and Song, Sibo and Dang, Kai and Wang, Peng and Wang, Shijie and Tang, Jun and Zhong, Humen and Zhu, Yuanzhi and Yang, Mingkun and Li, Zhaohai and Wan, Jianqiang and Wang, Pengfei and Ding, Wei and Fu, Zheren and Xu, Yiheng and Ye, Jiabo and Zhang, Xi and Xie, Tianbao and Cheng, Zesen and Zhang, Hang and Yang, Zhibo and Xu, Haiyang and Lin, Junyang},
  journal={arXiv preprint arXiv:2502.13923},
  year={2025}
}

@article{Qwen2VL,
  title={Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution},
  author={Wang, Peng and Bai, Shuai and Tan, Sinan and Wang, Shijie and Fan, Zhihao and Bai, Jinze and Chen, Keqin and Liu, Xuejing and Wang, Jialin and Ge, Wenbin and Fan, Yang and Dang, Kai and Du, Mengfei and Ren, Xuancheng and Men, Rui and Liu, Dayiheng and Zhou, Chang and Zhou, Jingren and Lin, Junyang},
  journal={arXiv preprint arXiv:2409.12191},
  year={2024}
}

@article{Qwen-VL,
  title={Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond},
  author={Bai, Jinze and Bai, Shuai and Yang, Shusheng and Wang, Shijie and Tan, Sinan and Wang, Peng and Lin, Junyang and Zhou, Chang and Zhou, Jingren},
  journal={arXiv preprint arXiv:2308.12966},
  year={2023}
}