Dreamina MCP Server
November 20, 2026
Table of contents
- Endpoint and authorization
- Setup
- Use it in the Claude app or ChatGPT
- Tools
- Long-running work
- Example prompts
- Limits and notes
The useapi.net Dreamina MCP server lets an AI assistant such as the Claude app, ChatGPT, Claude Code, Cursor or VS Code Copilot make videos and images with Dreamina on your own Dreamina accounts. It is a hosted Model Context Protocol server with 15 tools: generate Seedance 2.5, 2.0, HappyHorse and Wan videos from text, start and end frames or omni references, upscale them, raise their frame rate, turn a Seedance 2.5 draft into a full-quality video, generate Seedream, GPT Image and Nano Banana images, upload your own images, videos and audio, and clean up your generation history. Each tool calls the Dreamina API, so the same rules and limits apply. Nothing needs to be installed or hosted.
Is there a Dreamina MCP server? Yes. useapi.net, a third-party service not affiliated with Dreamina or ByteDance, hosts one at https://api.useapi.net/v1/dreamina/mcp with 15 tools for Seedance video and Seedream images on your own Dreamina plan, including its Unlimited mode, nothing to install.
Prefer REST? Code: github.com/useapi/dreamina-api
See all useapi.net MCP servers on Model Matrix & MCP.
Endpoint and authorization
https://api.useapi.net/v1/dreamina/mcp
- Transport: Streamable HTTP.
- Listed in the official MCP Registry as
net.useapi/dreamina, so MCP clients and directories that read the registry can find it by that name. - Authorization: your useapi.net API token in the
Authorization: Bearer …header, see Setup useapi.net. The Claude app and ChatGPT sign in with OAuth instead, see Use it in the Claude app or ChatGPT. - You need at least one Dreamina account added with Setup Dreamina. Adding and removing accounts is not available over MCP: use POST /accounts or the setup page.
Setup
Replace YOUR_API_TOKEN with your useapi.net API token.
Claude Code
claude mcp add --transport http useapi-dreamina https://api.useapi.net/v1/dreamina/mcp --header "Authorization: Bearer YOUR_API_TOKEN"
Cursor
In ~/.cursor/mcp.json:
{
"mcpServers": {
"useapi-dreamina": {
"url": "https://api.useapi.net/v1/dreamina/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_TOKEN"
}
}
}
}
VS Code
In .vscode/mcp.json:
{
"servers": {
"useapi-dreamina": {
"type": "http",
"url": "https://api.useapi.net/v1/dreamina/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_TOKEN"
}
}
}
}
Windsurf and other clients
Any client that supports remote MCP servers over HTTP with custom headers works. The usual JSON shape:
{
"mcpServers": {
"useapi-dreamina": {
"serverUrl": "https://api.useapi.net/v1/dreamina/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_TOKEN"
}
}
}
}
Some clients name the address field url instead of serverUrl.
Test it with curl
curl -s https://api.useapi.net/v1/dreamina/mcp \
-H "Authorization: Bearer YOUR_API_TOKEN" \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18","capabilities":{},"clientInfo":{"name":"curl","version":"1"}}}'
curl -s https://api.useapi.net/v1/dreamina/mcp \
-H "Authorization: Bearer YOUR_API_TOKEN" \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","id":2,"method":"tools/list"}'
The first call answers with the server’s name useapi-dreamina and its instructions for the assistant, the second with the 15 tools.
Use it in the Claude app or ChatGPT
The Claude app (claude.ai and Claude Desktop) and ChatGPT can add this server as a custom connector. They sign in with OAuth: you give the app the URL, it opens a useapi.net page, and you paste your useapi.net API token there once. Both work on their Free plans. The API-token setups above for Claude Code, Cursor, VS Code and other clients keep working as before.
https://api.useapi.net/v1/dreamina/mcp
Claude app
- Open Settings → Connectors and click Add custom connector.
- Enter a name, for example
useapi Dreamina, and the URL above, then click Add. - Click Connect. Claude opens the useapi.net sign-in page.
- Paste your useapi.net API token and click Allow.
Claude asks before each tool call unless you choose Always allow. The Free plan allows one custom connector.
ChatGPT
- Open Plugins, click Add and choose Add custom MCP server.
- Enter a name and the URL above, and set Authentication to OAuth.
- Tick ChatGPT’s “I understand and want to continue” box, then click Create as a plugin.
- Click Connect. ChatGPT opens the useapi.net sign-in page.
- Paste your useapi.net API token and click Allow.
ChatGPT asks you to confirm each tool call that creates or changes something. The Free plan works without developer mode (as of October 25, 2026).
How the sign-in works
- Your API token is in your welcome email, see Setup useapi.net. The useapi.net page links there too.
- Each connection is for one MCP server. Connect each server you want to use on its own.
- The app gets an access token that lasts 1 hour and renews it on its own. A connection you don’t use for 90 days lapses, and you connect again.
- Resetting your API token disconnects every app connected with it.
- Results come back as public links, so they open right in the app.
Tools
Accounts are named by their key, the region and the email, such as US:user@example.com. A job with references runs on the account that holds them, and upscale, frame rate and full quality run on the source job’s account. Tools that start a job take an optional account; without it a random account with a free slot whose region offers the model is picked.
Accounts
| Tool | What it does | Main parameters |
|---|---|---|
dreamina_list_accounts | Lists your accounts with their region, maxJobs and session expiry, without calling Dreamina. | |
dreamina_get_account | Checks one account live: the video and image models its region and plan offer, and its credits. Called once a day, it also claims the account’s free daily credits. | account |
Videos
| Tool | What it does | Main parameters |
|---|---|---|
dreamina_create_video | Generates a video with Seedance 2.5, 2.5 Flash, 2.5 Draft, 2.0, 2.0 Fast, 2.0 Mini, 1.x, HappyHorse 1.1 or Wan 3.0: from text, a first and last frame, 2 to 10 keyframes, omni references cited in the prompt as @image1, @video1 or @audio1, or a re-cut of your own video. unlimited: true uses Dreamina’s Unlimited mode at no credits. | prompt, model, ratio, duration, resolution, unlimited, firstFrameRef, endFrameRef, frames, omniImages, omniVideos, omniAudios, editVideoRef, account |
dreamina_get_video | Reads a video job, and with wait: true waits for it to finish. A finished video comes with its videoUrl, cover, size, duration and assetId. | jobid, wait |
dreamina_upscale_video | Doubles a finished video’s width and height, for example 1080p to 4K. Each video can be upscaled once. | jobid |
dreamina_interpolate_video | Raises a finished video’s frame rate to 30 or 60 fps. | jobid, fps |
dreamina_full_quality_video | Turns a finished seedance-2.5-draft video of up to 14 days old into a 1080p video, or 4K on a CA account. | jobid, resolution |
The parameters match POST /videos, POST /videos/upscale, POST /videos/interpolate and POST /videos/full-quality. The numbered references of the REST API (omni_N_imageRef, frame_N_imageRef and so on) are lists here: the first item of omniImages is @image1 in the prompt, the second @image2.
Images
| Tool | What it does | Main parameters |
|---|---|---|
dreamina_create_image | Generates 1 to 8 images, by model, with Seedream, GPT Image or Nano Banana, from a prompt and up to 6 reference images. The first image comes back as an image the assistant can see. | prompt, model, ratio, resolution, count, background, imageRefs, imageStrength, unlimited, account |
dreamina_get_image | Reads an image job, and with wait: true waits for it to finish. Each image comes with its URL, its assetRef to reuse as a reference and its assetId. | jobid, wait |
dreamina_upscale_image | Upscales one of a finished job’s first 4 images to 2K, 4K or 8K. | jobid, imageIndex, upscaleResolution |
The parameters match POST /images and POST /images/upscale.
Files, history and running jobs
| Tool | What it does | Main parameters |
|---|---|---|
dreamina_upload_asset | Uploads your own image (JPEG, PNG or WebP, up to 10 MB), video (MP4, MOV or WebM, up to 50 MB) or audio (MP3, WAV, M4A or AAC, up to 15 MB) from a public https URL, or sent as contentBase64, and returns an assetRef that works on that account only. Uploads cannot be deleted. | account, url or contentBase64, mimeType |
dreamina_list_assets | Lists an account’s generation history, images and videos, newest first, with each item’s assetId, page by page. | account, count, offset |
dreamina_delete_asset | Deletes a generated image or video from the account’s Dreamina history. It cannot be undone, and no credits come back. | assetId |
dreamina_list_running_jobs | Lists the jobs running now, by account. | |
dreamina_cancel_job | Stops tracking a running job and frees its slot. A job not yet started is marked failed. Credits already spent are not refunded. | jobid |
The full input schema of every tool comes from tools/list.
Long-running work
MCP clients usually give up on a tool call after about a minute, so a tool waits up to about 50 seconds:
- A quick image can come back from
dreamina_create_imageitself. - Most videos and slower images take longer. The tool then returns the
jobid, anddreamina_get_videoordreamina_get_imagewithwait: truepicks it up. Ask the assistant to keep checking until it is done.
Clients that ask for progress notifications get one about every 10 seconds while a tool waits.
Videos and images come back as Dreamina’s own download links, which open without sign-in. They expire, videos after about 7 days, so download them soon.
Example prompts
- “Make a 10-second Seedance 2.0 video of a paper boat drifting down a rainy street at night, 9:16.”
- “Upload this photo and this song, then make a 15-second Seedance 2.5 video where @image1 dances to @audio1.”
- “Make a Seedance 2.5 draft of this prompt, and if it looks right, render it in full quality at 1080p.”
- “Generate 4 Seedream images of a cozy mountain cabin in winter, then upscale the second one to 4K.”
- “Which video models and how many credits does my CA account have?”
- “Make the last video smoother at 60 fps.”
- “Delete the images we made today from my Dreamina history.”
Limits and notes
- The tools use your Dreamina accounts and the same rules as the REST API, such as
maxJobsand the models of each region. Generations use the account’s Dreamina credits, see Pricing, andunlimited: trueuses none on a plan that includes Unlimited, see Unlimited. A failed job is not charged. - A client may send up to 20 JSON-RPC requests in one batch.
- If the client sends an
MCP-Protocol-Versionheader, it must be one of the supported versions:2025-11-25,2025-06-18,2025-03-26or2024-11-05. - Tool calls show in your usage stats (
bot=dreamina) through the REST call each tool makes, the same rows a direct API request writes.