AI Video MCP Server for Image and Video Generation
Add the Auteo AI video MCP to Claude and turn a chat into finished videos, images and songs with Seedance, Nano Banana, GPT Image and Suno V6, without leaving the conversation.
- Create API key
Step 1
Create an API key
The AI video MCP signs in with an API key from your Auteo account. Generations spend the same credits as the website.
Step 2
Add the connector in Claude
Open the prefilled form: the name Auteo and the MCP server URL are already filled in. Under Request headers, add the header below with your key, then save.
Add Auteo to Claude(opens in a new tab)Header name
AuthorizationHeader value
Bearer YOUR_API_KEYStep 3
Ask for your first video
Try: “Check my Auteo credits, then make a 5-second Seedance 2.5 clip of a paper boat drifting down a rainy street.”
- Create API key
Step 1
Create an API key
The AI video MCP signs in with an API key from your Auteo account. Generations spend the same credits as the website.
Step 2
Add the server to Cursor
Paste this into ~/.cursor/mcp.json (or .cursor/mcp.json inside a project), replace YOUR_API_KEY with your key and reload Cursor.
Cursor
{ "mcpServers": { "auteo": { "url": "https://auteo.ai/api/mcp", "headers": { "Authorization": "Bearer YOUR_API_KEY" } } } }Step 3
Ask for your first video
Try: “Check my Auteo credits, then make a 5-second Seedance 2.5 clip of a paper boat drifting down a rainy street.”
- Create API key
Step 1
Create an API key
The AI video MCP signs in with an API key from your Auteo account. Generations spend the same credits as the website.
Step 2
Add the server to Claude Code
Run this once in your terminal and replace YOUR_API_KEY with your key. Claude Code sends the header on every request.
Claude Code
claude mcp add --transport http auteo https://auteo.ai/api/mcp \ --header "Authorization: Bearer YOUR_API_KEY"Step 3
Ask for your first video
Try: “Check my Auteo credits, then make a 5-second Seedance 2.5 clip of a paper boat drifting down a rainy street.”
- Create API key
Step 1
Create an API key
The AI video MCP signs in with an API key from your Auteo account. Generations spend the same credits as the website.
Step 2
Add the server to Codex
Add this to ~/.codex/config.toml, then export your key in the shell that starts Codex. Codex reads the variable and sends it as a Bearer token.
Codex
# ~/.codex/config.toml [mcp_servers.auteo] url = "https://auteo.ai/api/mcp" bearer_token_env_var = "AUTEO_API_KEY" # shell export AUTEO_API_KEY="YOUR_API_KEY"Step 3
Ask for your first video
Try: “Check my Auteo credits, then make a 5-second Seedance 2.5 clip of a paper boat drifting down a rainy street.”
MCP server URL
https://auteo.ai/api/mcpExample workflows with the AI video MCP
Once the AI video MCP is connected, you describe the result and your assistant picks the model, checks the price and brings the files back. A few things people ask for on day one:
Storyboard to video
Draft a product storyboard in chat, generate a still for every shot with Nano Banana, then animate the best frames with Seedance 2.5 image-to-video.
“Make four 16:9 storyboard frames for a coffee brand ad, then turn frame 2 into a 5-second video.”
Social clips in batches
Write a week of short vertical videos for TikTok and Reels. The AI video MCP estimates the cost before each clip, so you approve the spend once.
“Estimate the cost of seven 9:16 clips at 720p, 5 seconds each, then generate them.”
A soundtrack for every clip
Ask the AI video MCP for a short instrumental in the mood of your video, or have Suno V6 sing your own lyrics, then pair the song with the clip.
“Make an upbeat lo-fi instrumental titled “Morning Run” for the sneaker video.”
Assets inside your code
In Cursor, Claude Code or Codex, the AI video MCP generates hero images and background loops while you build the page that uses them.
“Generate a 1920×1080 abstract hero image in our brand blue and save the link in the README.”
What the Auteo MCP server for image and video generation can do
The same models and prices as auteo.ai, exposed as tools your assistant can call. Every tool works the same way in Claude, Cursor, Claude Code and Codex.
Text-to-video and image-to-video
Create clips from a prompt, from a first frame, or from first and last frames. Pick the duration, aspect ratio and resolution the model supports.
Text-to-image and image editing
Generate images up to 4K, or pass reference images to edit, restyle or keep a character consistent across shots.
Cost estimates before you spend
The AI video MCP prices every request in credits first, then waits for the job and brings the download links back into the chat, even after a timeout.
Music with lyrics or instrumentals
Describe a song and let the model write it, or pass your own lyrics, style and title. Ask for an instrumental or pick a male or female voice.
Video, image and music models in the AI video MCP
The AI video MCP serves the same video, image and music models as the Auteo studio. Your assistant reads the live list and each model's settings before it generates, so new models show up without changing your setup.
Video models
- Seedance 2.5
Clips up to 30 seconds at up to 1080P
- MiniMax H3
2K video with native stereo audio
- Seedance 2.0
Cinematic motion with output up to 4K
- Seedance 2.0 Fast
Quicker, lower-cost drafts at 480P–720P
Image models
- GPT Image 2.6
OpenAI's newest image model, Flare and Sunburst
- Seedream 5.5
ByteDance's newest image model, transparent PNG and stable edits
- GPT Image 2.5 Sunburst
Exceptional quality, precise edits
- GPT Image 2.5 Flare
Stunning everyday images, fast
- Nano Banana 2.1
Google's next-gen flagship image model
- Seedream 5.0 Pro
Logically consistent images with intelligent visual reasoning
- Nano Banana Pro
Google's flagship generation model
- Nano Banana
Google's standard generation model
- Nano Banana 2
High-speed image generation at scale
- GPT Image 2
Near-perfect text and extreme precision
- Z Image
Text-to-image generation at 1K or 2K
- FLUX.2
Text-to-image and reference-image editing at 1K or 2K
Music models
- Suno V6
More natural vocals and richer musical detail
- Suno V6 Mini
Lightweight and fast, quality with quick turnaround
- Suno V6 Wild
Bolder, more distinctive musical expression
- YuE2
Free and unlimited: open-source full songs, one per run
How the AI video MCP works
Auteo runs a remote MCP server for image and video generation over Streamable HTTP. Nothing to install: your client calls it with your API key.
01
Your assistant reads the tools
After you connect, the client lists the Auteo tools: account status, model list, model settings, cost estimate, upload, generate image, generate video, generate music and results.
02
It plans and prices the job
The assistant picks a model that fits your request, reads its valid settings and estimates the credits before it submits.
03
You get the files back
Videos usually take a few minutes. The assistant waits for the result and returns download links you can open, share or pass to the next step.
Why use the Auteo AI video MCP
One connection gives your assistant the image and video models you already use on auteo.ai, with the account you already have.
One account, one balance
No separate API billing. The AI video MCP spends your Auteo credits at website prices, and your plan's benefits apply.
Top models in one place
Seedance 2.5 and MiniMax H3 for video, Nano Banana and GPT Image for images, Suno V6 for music. Switch models in a sentence.
Works where you already work
Claude for chat, Cursor and Codex for code, Claude Code in the terminal. The same MCP server for image and video generation behind all of them.
You stay in control
Prices are shown before generation, keys can be revoked at any time in your profile, and every result also appears in your Auteo history.
AI video MCP FAQ
What is the Auteo AI video MCP?
Is the MCP server free?
Why do I need an API key?
Does it work with ChatGPT?
Which video, image and music models are included?
Can the remote MCP read files on my computer?
What happens if a generation times out?
How do I keep my API key safe?
What is the difference between the MCP and the CLI?
Can I make music through the MCP?
Do I need to be a developer to use it?
Connect the AI video MCP in two minutes
Create an API key, add Auteo to your client and ask for your first clip.