Qwen-Edit-2509-abliterated

🔄 Update Log / Changelog

v1.2 — 2025-12-16

Status: Released

Improvements

• 🎨 Improved anime-style consistency
• 🧍 More complete and anatomically correct male bodies
• 🧩 VAE / CLIP / UNet merged into a single checkpoint for easier deployment

Qwen-Edit-2509-Abliterated Model Card

Despite the existence of many unreviewed LoRAs and checkpoints based on qwen-image-edit, the image quality during actual usage is still not ideal. Especially for NSFW merge-model creation, adding too many LoRAs with arbitrary weights often damages the original semantic capability of qwen-image-edit.

Let’s introduce Qwen-Edit-2509-Abliterated.

What is it?

Qwen-Edit-2509-Abliterated is a merged model created by combining the Qwen-Edit-2509 checkpoint with various community LoRAs. It integrates multiple NSFW LoRAs, along with functional LoRAs such as camera-angle transformation, color tone reference, and image quality enhancement. The goal is to improve Qwen-Edit’s understanding of human poses and shooting perspectives during image editing.

As its name suggests, this is a model without review mechanisms. It can generate NSFW and unreviewed content.

The model creator is not responsible for any outputs produced. Users are encouraged to use this checkpoint friendly, legally, and responsibly.

The merged model has quantized in FP8.

The model has Supports 4-step accelerated sampling.

Some of my personal insights on LoRA merging

1. About accelerator selection

When comparing merge models with similar purposes, we found that most NSFW LoRAs were originally trained on qwen-image, a model designed for image generation. However, accelerators used for these merge models are often from qwen-image-edit.

Tests showed:

• qwen-image-edit accelerators often fail during NSFW generation
• This may result in incomplete characters or “broken images”
• Recent testing suggests that the qwen-image accelerator performs much better for LoRA acceleration than the native qwen-image-edit one

Therefore, v1.0 uses the qwen-image accelerator, though this still requires further validation.

2. About LoRA weight selection

The LoRA weights used in merging are all kept small (usually 0.2–0.3). This seems to preserve most of qwen-image-edit’s original semantic understanding.

In actual usage, the experience has been very good.

Have fun!