Qwen-Image-Edit-2511-LoRAs-Fast

Tools

• QWEN_EDIT_IMAGE_api_pose_still Letsnude still: copy pose from a dataset frame onto the user photo. image : user start photo (identity + background to keep). pose_ref : still from a train clip already in the sex pose (penetration visible). prompt : optional extra instruction; blank uses the AnyPose copy-pose prompt. lock_face_hair : segmentor pastes original face+hair so expression does not follow pose_ref. Returns : [result_image, seed_used] — feed this still into kv-i2v as first/last frame. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
• QWEN_EDIT_IMAGE_api_predict Edit an image from a text prompt (Qwen-Image-Edit-2511). Returns [result_image, seed_used]. image : the image to edit (the person/subject whose identity is kept). prompt : what to change, e.g. "change her outfit to a red dress; turn to look left". lora_adapter : optional style name from the adapter list ("" = no style LoRA / base edit). image2 : optional second reference image (e.g. a style or lighting reference). seed : int seed (ignored if randomize_seed=True). randomize_seed : pick a new random seed each call. guidance_scale : true-CFG. 1.0 = fastest; >1 (e.g. 2.5) activates the negative prompt that suppresses extra fingers / identity drift (slower). steps : denoise steps. 4 is native for this distilled model; pose still uses 8. preserve_identity : keep the person's face while allowing pose/head to change. identity_strength : 0-100, how forcefully identity is locked. output_size : long-side pixels 1024-2048. Higher = more HD

No tool matches.

As the server's own listing gives them. We never call a tool.