off-grid-ai/Qwen-Image-2.1-GGUF

Qwen Image 2.1 — GGUF

GGUF conversions of Qwen Image 2.1 for on-device image generation with stable-diffusion.cpp. Converted with sd.cpp's -M convert.

Files

• Qwen-Image-2.1-Q4_K_S.gguf — smaller Q4 quantization.
• Qwen-Image-2.1-Q8_0.gguf — Q8 quantization.

Q4_K_S in a file name: stable-diffusion.cpp wrote the file with --type q4_K. Its weight tensors are Q4_K and the other tensors stay F16/BF16/F32. Q4_K_S is the closest standard GGUF label, so Hugging Face can list the file.

Required companion files, from off-grid-ai/Qwen-Image-2.1-Companions-GGUF:

• Qwen3VL-8B-Instruct-Q4_K_M.gguf — Qwen3-VL 8B text encoder, 5.028 GB, Apache-2.0.
• qwen_image_2.1_vae_bf16.safetensors — Qwen Image 2.1 VAE, 0.676 GB, Qwen Research License.

The same companion files serve Qwen Image 2.1 and Qwen Image 2.1 Turbo.

Q4 optimized uses the same Q4 file with --fa --vae-conv-direct --vae-tiling.

Use with stable-diffusion.cpp

Download one generator and the companion files into one folder. Use a stable-diffusion.cpp build that supports this model.

hf download off-grid-ai/Qwen-Image-2.1-GGUF Qwen-Image-2.1-Q4_K_S.gguf --local-dir .
hf download off-grid-ai/Qwen-Image-2.1-GGUF --include 'run_example.py' 'benchmarks/*' --local-dir .
hf download off-grid-ai/Qwen-Image-2.1-Companions-GGUF Qwen3VL-8B-Instruct-Q4_K_M.gguf qwen_image_2.1_vae_bf16.safetensors --local-dir .
python run_example.py --scene tabletop --variant q4_optimized --engine /path/to/sd-cli

Examples, measurements, and full settings. Timings are recorded runs, not controlled speed comparisons; shared GPU work can affect them.

Earlier file names

Files were renamed on 2026-10-21 so that Hugging Face lists every quantization and does not read a companion file as the model. File contents are unchanged (same SHA-256). Links pinned to an earlier commit still work.

Earlier name	Current name
base-q4_K.gguf	Qwen-Image-2.1-Q4_K_S.gguf
base-q8_0.gguf	Qwen-Image-2.1-Q8_0.gguf
Qwen3VL-8B-Instruct-Q4_K_M.gguf	moved unchanged to off-grid-ai/Qwen-Image-2.1-Companions-GGUF
qwen_image_2.1_vae_bf16.safetensors	moved unchanged to off-grid-ai/Qwen-Image-2.1-Companions-GGUF

Credit & license

Original model: Qwen/Qwen-Image-2.1 — created by its respective authors.

License: qwen-research (carried over from the original; its use restrictions apply). See NOTICE for attribution and conversion notices.

This is a format conversion (quantization) only; all credit for the model belongs to the original creators.

Improved using Qwen.