Sora API and MCP: how to add video generation to your project

Sora is a family of video generation models. NeuralSpace offers the following models: Sora 2, Sora 2 Pro. They all run through a single public API, and there is a ready MCP server for AI agents. The key is issued on the API keys page. Below are the model capabilities, request examples and setup steps.

What Sora can do

Sora 2 generates video from text, priced by duration (10 or 15 seconds). Sora 2 Pro additionally offers standard or high quality. Landscape and portrait orientations are supported, and image_urls lets you pass a source image for image-to-video.

API endpoint

Send a request to POST /v1/videos/generations. Example: {"model": "sora-2", "prompt": "waves on the shore at sunset", "duration": 10, "aspect_ratio": "landscape"}. Authorization uses an API key in the x-api-key header. Available model ids: sora-2, sora-2-pro.

MCP tools

The NeuralSpace MCP server is available at POST /mcp (Streamable HTTP). To work with Sora, an agent uses the tools: generate_video, check_video, list_models, get_balance. This lets Claude Code, Cursor and other MCP clients use the site models directly.

How to connect

Sign up on NeuralSpace, open the API keys page and create an nsk-… key. Then pass it in the request header or in your MCP client settings. Full documentation is on the the /v1/docs page.

FAQ

Which durations are available?

10 and 15 seconds.

How does Pro differ from the standard version?

Pro additionally lets you choose standard or high quality.

Does it support references?

Yes, image-to-video via image_urls.