Qwen-Rapid-AIO-NSFW-v11
Back

Qwen-Rapid-AIO-NSFW-v11
11.3K 6 74

ACG design

Qwen-Rapid-AIO-NSFW-v11

Combine accelerators, VAE, and CLIP to easily and quickly support Qwen image editing (and text-to-image) functionality.

Use the "Load Checkpoint" node. 1 CFG, 4 steps. Use the "TextEncodeQwenImageEditPlus" node to process input images (optional) and prompts. If no image is provided, it performs pure text-to-image conversion. FP8 precision.

Both NSFW and SFW models are provided! v4 and earlier versions combine NSFW and SFW purposes into one model, but with poorer performance. Starting from v5, NSFW and SFW versions are separated. Please choose the appropriate model based on your use case.

Having issues with scaling, cropping, or resizing? The problem lies in scaling images within the TextEncoderQwenEditPlus node. There are many solutions, but I prefer directly fixing the node, and I have provided my version in the file section. It also supports up to 4 input images. Set the "target_size" to slightly smaller than the maximum output size (e.g., set to 896 when creating a 1024x1024 image). I found this improves quality better than skipping scaling entirely, as the input images better match the output resolution.

This model is sourced from an external transfer (transfer address: https://huggingface.co/Phr00t/Qwen-Image-Edit-Rapid-AIO ）,if the original author has objections to this transfer, you can click，
Appeal
We will, within 24 hours, edit, delete, or transfer the model to the original author according to the original author's request

user_iirk1dai

user_iirk1dai

ACG design

Model Information

Active
Original author：
Phr00t
Model Type：
Checkpoint
Basic Model：
Qwen-Edit-2509
Resource Name：
models/checkpoints/Qwen-Rapid-AIO-NSFW-v11.safetensors
MD5：
6e65c9ad43239a97e9ddf1427b53bc81

Combine accelerators, VAE, and CLIP to easily and quickly support Qwen image editing (and text-to-image) functionality.

Use the "Load Checkpoint" node. 1 CFG, 4 steps. Use the "TextEncodeQwenImageEditPlus" node to process input images (optional) and prompts. If no image is provided, it performs pure text-to-image conversion. FP8 precision.

Both NSFW and SFW models are provided! v4 and earlier versions combine NSFW and SFW purposes into one model, but with poorer performance. Starting from v5, NSFW and SFW versions are separated. Please choose the appropriate model based on your use case.

Having issues with scaling, cropping, or resizing? The problem lies in scaling images within the TextEncoderQwenEditPlus node. There are many solutions, but I prefer directly fixing the node, and I have provided my version in the file section. It also supports up to 4 input images. Set the "target_size" to slightly smaller than the maximum output size (e.g., set to 896 when creating a 1024x1024 image). I found this improves quality better than skipping scaling entirely, as the input images better match the output resolution.

This model is sourced from an external transfer (transfer address: https://huggingface.co/Phr00t/Qwen-Image-Edit-Rapid-AIO ）,if the original author has objections to this transfer, you can click，
Appeal
We will, within 24 hours, edit, delete, or transfer the model to the original author according to the original author's request