Model Name:

qwen_image_edit_2511_fp8_e4m3fn_scaled_lightning.safetensor

Model Type:

FP8 (e4m3fn scaled) precision, fused with a 4-step distilled LoRA, optimized for low-memory deployment

Feature Showcase

Qwen-Image-Edit-2511 Enhanced Character Consistency

In Qwen-Image-Edit-2511, character consistency has been significantly improved. The model can perform imaginative edits based on an input portrait while preserving the subject's identity and visual features.

Improved Multi-Character Consistency

While Qwen-Image-Edit-2509 improved consistency for single-character editing, Qwen-Image-Edit-2511 further enhances consistency in multi-person group photos—blending two separate individual images into a coherent group photo with high fidelity:

Built-in Support for Community-Created LoRAs

Since the release of Qwen-Image-Edit, the community has created numerous creative and high-quality LoRAs, greatly expanding its expressive potential. Qwen-Image-Edit-2511 directly integrates selected popular LoRAs into the base model, delivering their effects right out of the box without requiring extra parameter tuning.

For example, the Lighting Enhancement LoRA now enables realistic lighting control out of the box:

As another example, generating new perspective views can now be handled directly by the base model:

Industrial Design Applications

We paid special attention to practical engineering scenarios—for instance, batch industrial product design:

...and material replacement for industrial components:

Qwen-Image-Edit-2511 introduces stronger geometric reasoning capabilities—for example, it can directly generate auxiliary construction lines for design or annotation purposes:

That covers the key updates in Qwen-Image-Edit-2511. Have fun exploring these new capabilities!