Bailian CLI command reference for vision understanding — image, document, and long video parsing with multilingual OCR.
A single command invokes Bailian's vision understanding capabilities from the terminal — parsing images, documents, and long videos.
ImportantAppend --help to any command for full parameter details, e.g. bl vision describe --help. Every parameter that accepts a file URL also accepts a local path — the CLI auto-uploads it to temporary storage (valid for 48 hours), so you never need to host files as URLs yourself. Command details on this page are verified against bl 2.2.1.
Command reference
bl vision describe
Describe an image or video using Qwen-VL. Requires API Key authentication.
Usage
bl vision describe --image <path-or-url> [--video <url>] [--prompt <text>]
| Parameter | Type | Required | Description |
|---|---|---|---|
--image <path-or-url> | string | no | Local image path or URL |
--video <url> | array | no | Video file URL or local path (mp4/mov/avi/mkv/webm) |
--prompt <text> | string | no | Question about the content (default: auto-detected) |
--model <model> | string | no | Vision model (default: qwen3-vl-plus) |
Example: Describe image using Qwen-VL
bl vision describe --image photo.jpg