AI Agent Hub
Back to skills
Jimeng Image-to-Image CLI icon

Jimeng Image-to-Image CLI

Design & Media Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md to install @user_c1d16043/jimeng-cli-image2image.

About this skill

Problem

When an agent already has local reference images, text-to-image or video workflows do not fit image-to-image editing. This skill turns Jimeng CLI image editing into an executable parameter contract:
- --images accepts 1–10 readable local files, comma-separated when multiple.
- Each submission must explicitly pass --resolution_type=1k|2k|4k.
- Model, resolution, and size combinations should be validated with dreamina image2image -h.

Workflow

Treat help output as the source of truth for the current CLI, then run the async generation loop:
- Validate that input images are readable, check credits and consumption impact, submit the job, and save submit_id.
- For querying states, keep calling query_result until success or fail.
- When using --width and --height for custom sizes, both must be positive integers and --ratio should be omitted; generate_num is limited to 1–10.

Boundaries

This skill is not for text-to-image, video generation, or pure prompt polishing.
- 1k image-to-image is supported only by 5.0Pro; 4.x and 5.0 support 2k or 4k only.
- A successful submit is not a successful generation.
- When the final state fails, report fail_reason.

Use Cases

  • A designer receives three local product reference images, runs Jimeng CLI image-to-image, and polls until success or failure.
  • An agent receives user-uploaded local images, submits an explicit 2k or 4k image-to-image job, saves submit_id, and polls the final state.
  • A content team generates variants from five to ten local assets, sets paired width/height and 1-10 generate_num, then reads fail_reason.
  • An engineer debugs v1.4.14 parameter combinations and uses help to validate model, resolution, and custom size before submission.

Best For

  • Engineers maintaining agent workflows: turn local reference images into image-to-image jobs and reliably track state.
  • Design tool developers: embed Jimeng CLI parameter constraints into multimodal generation pipelines.
  • AIGC content operations: batch-generate controlled-resolution images from 1-10 local assets.
  • CLI integration test engineers: verify failure causes for model, resolution, and width/height parameter combinations.