AI Agent Hub
Back to skills
Lux3D China 3D Generation API icon

Lux3D China 3D Generation API

Design & Media Updated 2026.08.30

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

Follow https://skillhub.cn/install/skillhub.md and install @user_97275c6e/lux3d-cn.

About this skill

Problem

3D asset generation APIs often scatter creation, polling, format selection, and result extraction across several HTTP calls. Local files, multi-view images, version parameters, and output formats are easy to wire incorrectly. Lux3d-cn wraps the Lux3D China API through lux3d_client.py, which is useful for engineers who want to script or automate image-to-3D, text-to-3D, material redraw, and export workflows.

How it works

  • Prerequisites: img, imgs, meshUrl, and modelUrl must be HTTP(S) URLs; the China API key is not compatible with international endpoints, and the authorization header passes the API key directly.
  • Core capabilities: image supports G1 and G1-Turbo; text supports prompt, style, and one reference image; material uses v3.0-standard to redraw GLB materials; four-view produces additional viewpoint images; export converts among zip, glb, usdz, obj_zip, and fbx_zip.
  • Task flow: creation endpoints return a taskid; use query or Python query_task_status to poll. On success, read URLs from d.outputs[].content and save them promptly because they expire after 2 hours.
  • Parameter boundaries: img and imgs are mutually exclusive; imgs accepts 1-32 images of the same subject; faceCount ranges from 10000 to 300000; enablePbr applies only to G1-Turbo or text-to-3D Turbo requests that include ZIP/GLB outputs.

Boundaries

This skill targets the Lux3D China environment and does not add business types or internal versions. Four-view generation is an optional enhancement path, not a prerequisite for G1-Turbo; export ZIP input must come from Lux3D generated tasks. Free and Pro accounts have 2 and 5 concurrent tasks respectively, so wait for existing tasks when concurrency limits are hit.

Use Cases

  • E-commerce operators with product photos need batch GLB/ZIP 3D assets for preview and download.
  • Designers with existing GLB models need material redraw from a reference image and USDZ export.
  • Game planners with text prompts need PBR or white-clay concept models for early validation.
  • Engineers need scripts to poll task status and extract result URLs from d.outputs[].content.

Best For

  • E-commerce operators who need product photos converted into 3D preview assets and common exports
  • 3D designers who have GLB models and want material redraw from reference images for AR output
  • Game planners who want text-described concept models with face-count and PBR options
  • Python engineers integrating the Lux3D China API into automated pipelines