Runway plugs its video models straight into Claude Opus 5.5 via MCP

Runway has integrated its video and image models into the Claude Opus 5.5 workspace using the Model Context Protocol. The connection lets users generate visual content directly inside the Claude chat, without switching between separate interfaces.

Models exposed through the integration

The MCP layer surfaces Runway's own Gen-4.5, Seedance 2.5, GPT Image 2 and Kling. A user can ask Claude to write a prompt, send it to the appropriate model, and receive a finished video or image back in the same conversation.

What MCP is and why it matters

MCP is an open standard that lets language models call external tools in a structured way. Until now, using Runway's video models meant logging into Runway's platform, pasting prompts manually, and downloading the output. The MCP connection folds that process into the natural Claude workflow: write an instruction, and the model selects and invokes the right visual tool.

Availability

According to the announcement, the integration is live immediately and takes "seconds" via a link included in the release. Runway did not specify whether a particular Claude or Runway subscription tier is required, and there is no public technical documentation yet on rate limits, supported resolutions, or any extra costs beyond existing subscriptions.