MCP Glossary

What is MCP video generation?

MCP video generation lets an AI agent create video clips by calling a video tool over the Model Context Protocol. You describe the clip, the agent calls the MCP server, and a finished, hosted video comes back. The agent can also upscale, reframe, or remove the background of existing video.

Text-to-video and beyond

The AetherWave MCP server exposes video generation across multiple leading models, plus video upscaling, reframing for different aspect ratios, and background removal. The agent calls aetherwave_generate_video, and you can list available models and pricing with aetherwave_list_video_models.

Because video is the most expensive media to produce, MCP's structured pricing helps: the agent can report the credit cost before generating so there are no surprises.

Composing video with other media

The real power is composition. In one conversation an agent can generate a track, produce matching cover art, then create a short video using that art, all through tools on the same server with one credit pool.

Generated clips save to your AetherWave gallery on Cloudflare R2, so links stay live.

Frequently Asked Questions

What video models are available?

Multiple leading text-to-video and image-to-video models. Run aetherwave_list_video_models from your agent to see the live catalog with current pricing.

Can the agent reframe a video for vertical?

Yes. The reframe tool adapts a clip to different aspect ratios, useful for turning a wide video into a vertical short.

Try it yourself

AetherWave's MCP server brings AI music, image, and video generation into your agent. Free API key, free starter credits, one install.

Get Your Free API Key