cli

This is the official command-line tool for Aliyun Model Studio, which is Alibaba Cloud's AI platform also known as Bailian or DashScope. It lets you use the platform's AI capabilities directly from a terminal, and it is also designed so that AI agents can call it as a structured tool rather than a human typing commands manually.

The tool covers a wide range of AI features. Text chat lets you talk to Qwen, Alibaba's family of language models. Multimodal commands accept text alongside images, audio, and video. Image generation and video generation commands take a text description and produce visual output. Speech synthesis can convert text to audio, including cloning a voice from a short sample. Image understanding commands can parse charts, documents, and perform visual reasoning. There are also commands for searching the web in real time, managing a knowledge base that stores documents for retrieval, and calling agents or automated workflows that you have already published on the platform.

Installation is through npm and requires Node.js version 22.12 or newer. Authentication uses an API key from the DashScope console. A browser-based login is required separately for commands that interact with the Alibaba Cloud console, such as listing published apps or checking how much free-tier quota you have used. Configuration is stored in a file at ~/.bailian/config.json and can be adjusted with a config command.

The README includes a demonstration of generating a two-minute cinematic video from a single sentence by chaining this CLI with a separate scene-planning skill and a Qwen coding model. The README is written in English with a link to a Chinese version. The project is released under the Apache 2.0 license.