HeadsUpAI

Vercel Launches AI CLI to Give Agents Native Multimodal Generation Powers

· Updated

Vercel, the frontend cloud platform, launched AI CLI—a terminal interface for multimodal generation. Built on the Vercel AI SDK, the tool provides unified access to hundreds of models. It is an agent skill (a standardized package of capabilities for AI agents), allowing autonomous tools to install it via one command.

This release expands the agent-native infrastructure from Vercel Labs. It follows wterm's terminal rendering and agent-browser's web automation, shifting AI from chat interfaces into the developer's native environment. By exposing generation as a CLI utility, Vercel enables Unix-style piping (passing one command's output as another's input).

You can use ai-cli to compare multiple models side-by-side or generate assets without context-switching. The tool supports inline previews for images and video frames in compatible terminals using WebAssembly-based decoding. It is available now via npm install -g ai-cli, requiring an AI Gateway or provider-specific API key.

View the full update on github.com
Chris Tate
Chris Tate
@ctatedev
X

Introducing AI CLI Generate images, video, and text from your terminal. Pipe them together. Works with any agent. ai image "sunset" | ai video "animate" → Hundreds of models → Multi-model comparison → Inline previews, no native deps → AI SDK + AI Gateway https://t.co/9Nvq25JNNt

60retweets1klikes
View on X

Still wondering? A few quick answers below.

Vercel AI CLI is a command-line tool for generating text, images, and video directly from the terminal. It uses the Vercel AI SDK and AI Gateway to provide unified access to hundreds of different AI models. The tool supports Unix-style piping, which is the ability to pass the output of one command as input to another.

You can install the tool globally using npm by running the install command for ai-cli. To use it, you need an API key for Vercel AI Gateway or a specific provider like OpenAI. Commands like ai image or ai video generate content based on your prompts, which can be saved to files or piped to other utilities.

Yes, the tool allows you to generate content using multiple models simultaneously by passing a comma-separated list of model IDs to the model flag. This feature enables developers and researchers to compare results from different providers, such as OpenAI and Google, side-by-side for the same prompt to evaluate quality and performance.

Vercel AI CLI is designed to be agent-native and can be added to AI coding agents as a skill, which is a packaged capability for autonomous systems. This allows agents to install the tool and use its multimodal generation capabilities to create text, images, or video assets as part of their automated development workflows.

The tool automatically displays generated images and video frames directly in the terminal if you use a compatible terminal like Kitty or iTerm2. It uses the Kitty graphics protocol, a standard for rendering high-quality images in a terminal, and a WebAssembly-based decoder to show previews without requiring native dependencies or external software on your machine.

Share this update