license: apache-2.0
base_model:
- Phr00t/Qwen-Image-Edit-Rapid-AIO
- Qwen/Qwen-Image-Edit-2509
- lightx2v/Qwen-Image-Lightning
pipeline_tag: text-to-image
library_name: comfyUI
tags:
- qwen
- qwen-edit
- t2i
- i2i
Only quants of https://huggingface.co/Phr00t/Qwen-Image-Edit-Rapid-AIO
READ CAREFULLY!!!
I think I have fixed the smaller quants, feel free to test and give a feedback.
I left the warning anyway.
ATTENTION !!! The smaller quants, smaller than Q5 except Q4_K_M creates weird artifacts, see https://huggingface.co/QuantStack/Qwen-Image-Edit-2509-GGUF/discussions/6#68e7099a87b5c3907529d4bc
Not AIO, needs the text encoder and the VAE
I personally recommend:
text_encoder https://huggingface.co/mradermacher/Qwen2.5-VL-7B-Instruct-abliterated-GGUF/tree/main mmproj needs clip fix in ComfyUI-GGUF/loader.py line 13
TXT_ARCH_LIST = {"t5", "t5encoder", "llama", "qwen2vl", "clip"}
see https://github.com/city96/ComfyUI-GGUF/pull/349
I have cloned the abliberated files into my repo and fixed the mmproj, so the fix is no longer needed
VAE https://huggingface.co/calcuis/pig-vae/blob/main/pig_qwen_image_vae_fp32-f16.gguf
Replace comfy_extras/nodes_qwen.py with https://huggingface.co/Phr00t/Qwen-Image-Edit-Rapid-AIO/blob/main/fixed-textencode-node/nodes_qwen.py
workflow text2image 4-step Q2_K:
workflow edit of this image 4-step Q2_K:
workflow edit of this image 4-step Q5_K_M:
