Aliyun Bailian | 阿里云百炼Install

Summary

Edit one or two images by providing URLs and a text instruction using the selected Qwen image editing model, receiving one or more modified images as the output.

Inputs

• Model | 模型 (required) — Choose the Qwen image edit model to use for editing, options vary by capability and quality.
• Instruction | 编辑指令 (required) — Provide a text instruction describing the desired edits to apply to the input images.
• Input Images | 图像 URL 列表 (required) — List one or two image URLs (one per line) to be edited, ordered as referenced by the instruction.
• Negative Prompt | 反向提示词 — Optionally provide negative prompts describing elements to avoid in the edits.
• Image Count | 图片数量 — Specify how many edited images to generate. Only certain models support multiple images; the basic model supports only one.
• Advanced Options | 高级选项 — Configure options such as automatic prompt refinement, random seeds for reproducibility, watermark inclusion, and seed randomization.
• Poll Interval (MS) — Set the delay between polling attempts while waiting for the editing task to complete.
• Max Wait (Seconds) — Set the maximum duration to wait for task completion before timing out.

Output shape

a list of edited image data objects corresponding to the generated images

By default, the response includes one edited image; advanced models can return multiple images as specified by the Image Count input. Image output includes URLs or data suitable for downstream workflow use. Operation polls the task status until completion or timeout based on Poll Interval and Max Wait settings.

Links

• n8n community nodes installation guide