Image & Video Generation MCP for Developers
AIMCP is building an MCP server that will let any MCP-capable client generate images and videos from one credit balance, and it is coming soon.
Everything on this page is labelled Coming soon. Tool names are planned and may change before launch, and no endpoint has been published yet. The same models are coming to the AIMCP web generator, which opens soon.
Planned MCP tools
Coming soon: the first version is planned around three tools. Names and parameters are not final.
- Planned
generate_image
Create or edit an image from a prompt and optional reference images, with a chosen model, aspect ratio and resolution.
- Planned
generate_video
Start a text-to-video or image-to-video task with a chosen model, duration and resolution. Returns a task ID.
- Planned
get_task_status
Check a generation task by ID and get the result when it is finished.
Pricing (Coming soon)
MCP calls are planned to use the same AIMCP credits as the web generator, with the credit cost depending on the model and settings. Current credit packs and plans are on the pricing page.
Supported clients (Coming soon)
The server is designed for the Model Context Protocol, so it is planned to work with any MCP-capable client.
- Claude (custom connectors)
- ChatGPT
- Cursor
- Codex
- Other MCP-compatible clients and agents
How the AIMCP MCP server will work
The planned flow for developers. Every step is in Early access and the details are coming soon.
- 01Early access
Join Early access
Email the AIMCP team to join Early access. Connection details will be shared with Early access members; there is no public endpoint yet.
- 02Early access
Add the server to your MCP client
Register the AIMCP server in Claude, ChatGPT, Cursor, Codex or another MCP client using the details from your invite, then sign in with your AIMCP account.
- 03Early access
Call the tools
Call generate_image or generate_video to start a task, then call get_task_status with the task ID to fetch the finished file. Credits come from your AIMCP balance.
What developers are planning to build
Agent workflows
Let a coding or research agent produce images and short clips as part of a larger task.
Content pipelines
Generate social images and video drafts from a content calendar.
Prototype assets
Create placeholder art and motion for apps, games and landing pages from inside your IDE.
Product media
Render product shots and image-to-video clips from catalog photos.
Model comparison
Run the same prompt across several models and compare the results.
Internal tools
Give your team image and video generation inside the assistant they already use.
Developer FAQ
Is there a public AIMCP API or MCP endpoint?
+
Not yet. The MCP server is coming soon and no endpoint has been published. Early access members will receive connection details first.
Which tools will the MCP server have?
+
The plan is generate_image, generate_video and get_task_status. These names are planned and may change before launch.
How will authentication work?
+
Coming soon. You will connect with your AIMCP account, and the exact sign-in flow will be documented at launch.
How will MCP usage be billed?
+
MCP calls are planned to use the same AIMCP credits as the web generator. See the pricing page for current packs and plans.
Which models will be available?
+
The server is planned to expose the same image and video models as the AIMCP web generator.
How do I get notified at launch?
+
Join Early access by email. We will contact you when the MCP server opens.