Haverbex/Qwen-Image-2.1-UltraFast-GGUF

Qwen-Image-2.1-UltraFast GGUF

Q4_K · 3.77 GiB · Adapter already merged. The diffusion transformer for eight-step image-editing research. Built with Qwen · Non-commercial research.

Model comparison gallery (BF16) · Detailed usage and conversion report

Download

hf download Haverbex/Qwen-Image-2.1-UltraFast-GGUF \
  Qwen-Image-2.1-UltraFast-Q4_K.gguf \
  --local-dir ./models/diffusion_models

Use a runtime with Qwen-Image-2.1 support, its matching VAE, and the required text/vision encoders. Do not apply the same adapter again. Runtime setup and recorded schedule.

The export passed tensor checks and a byte-identical native read/write roundtrip. GGUF image quality and speed remain unbenchmarked. The linked gallery shows BF16 adapter outputs.

License · Notice · Checksums