abenzerps/Qwen-Image-2.1-Turbo-Quantized
AI model · https://huggingface.co/abenzerps/Qwen-Image-2.1-Turbo-Quantized
First seen by Oriel on .
AI summary
This Hugging Face repository provides quantized formats of Qwen-Image-2.1-Turbo and setup instructions for generating images with ComfyUI, stable-diffusion.cpp, or MFLUX.
- The base model is a 7B visual generation architecture using eight denoising steps and CFG 1; 1024 × 1024 generation takes about 10–20 seconds on GPU after warm-up.
- Available formats include GGUF, FP8 / INT8 ConvRot / NVFP4 Safetensors, and MLX 4-bit / 6-bit / 8-bit weights.
- ComfyUI and command-line setups require a diffusion model, a text encoder, and a VAE; the guide recommends Q4_K_M for GGUF and INT8 ConvRot for Safetensors.
- The Turbo sampler uses a specific nine-value sigma schedule for eight steps; a regular KSampler with simple scheduling does not reproduce it.
- The guide notes that 2048 × 2048 needs more memory and time, and that model file size does not represent total VRAM use.
#image generation #quantization #comfyui #mlx
Generated by gpt-5.6-sol. Check the original for details.
Metric trends
Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained
Likes
Total likes reported by the source.
- Hugging Face Models: 6,039, latest observation ; +416 over observed period since
Downloads (30d)
Rolling 30-day downloads. A change in this value is not new downloads during the observed period.
- Hugging Face Models: 165,081, latest observation ; -2,343,509 over observed period since
Aliases
| Source | External ID | URL |
|---|---|---|
| Hugging Face Models | abenzerps/Qwen-Image-2.1-Turbo-Quantized | https://huggingface.co/abenzerps/Qwen-Image-2.1-Turbo-Quantized Read discussion |