seedream-mcp
A focused MCP server for generating and editing images using Seedream AI models, with support for task management and pricing lookups.
Generate images from text (
text_to_image): Submit a text prompt to one of three Seedream text-to-image models (seedream-4.5,seedream-5-lite,seedream-v4) and receive output image URLs.Edit images (
edit_image): Submit source image URLs to a Seedream edit model to generate edited outputs.Configurable outputs: Set aspect ratio, output count (1–6), quality (basic/high), and resolution (1k/2k/4k) for both workflows.
Wait or fire-and-forget: Use the
waitparameter to poll until completion and get results immediately, or submit asynchronously and check back later.Poll task status (
get_task): Fetch the current status and result payload of a previously created task using its task ID.Check pricing (
check_pricing): Look up current pricing for any Seedream model and endpoint — no API key required, useful for cost estimation.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@seedream-mcp generate an image of a cat wearing a space helmet"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Why This Package?
@runapi.ai/seedream-mcp is a focused Model Context Protocol server for the Seedream model line on RunAPI.
It gives MCP-compatible assistants direct access to 3 endpoints and 9 model variants without loading the full RunAPI catalog.
Use this per-model server when an agent should stay scoped to Seedream. Use @runapi.ai/mcp when one assistant should discover every RunAPI model line.
Related MCP server: grok-imagine-mcp
Install
Add it to Claude Code:
claude mcp add seedream -s user -- npx -y @runapi.ai/seedream-mcpUse project scope when the server should be shared with a repository:
claude mcp add seedream -s project -- npx -y @runapi.ai/seedream-mcpCodex, Cursor, Windsurf, VS Code, Roo Code, and other MCP hosts can use the same stdio command:
{
"mcpServers": {
"seedream": {
"command": "npx",
"args": ["-y", "@runapi.ai/seedream-mcp"]
}
}
}check_pricing works before sign-in. For task creation and status polling, ask your assistant to call the login tool. It opens a browser login and saves credentials to ~/.config/runapi/config.json, the same file used by runapi login.
Headless and CI hosts can still set RUNAPI_API_KEY before starting the MCP host.
Ready-made examples are in examples/ for Claude, Cursor, Windsurf, VS Code, and Roo Code.
Tools
Tool | Auth | Purpose |
| Yes | Create a Seedream decompose layers task and optionally wait for a terminal status. Returns the task id, status, and output URLs. |
| Yes | Create a Seedream edit image task and optionally wait for a terminal status. Returns the task id, status, and output URLs. |
| Yes | Create a Seedream text to image task and optionally wait for a terminal status. Returns the task id, status, and output URLs. |
| Yes | Fetch the current status and latest payload for an existing task. |
| No | Look up current pricing for a Seedream model and endpoint. |
Models
Seedream covers 9 model variants across 3 endpoints. Each tool accepts the models listed for it:
Tool | Models |
|
|
|
|
|
|
Model availability can change between releases. Use check_pricing or the Seedream model page for the current catalog view.
Agent Prompts
Ask your assistant in natural language; it can inspect pricing, create the task, and return the task id plus output URLs.
Create a task
Run a Seedream decompose layers task with RunAPI.The assistant can call check_pricing, then decompose_layers, and return the task id, status, and output URLs.
Submit without waiting
Create the task but don't wait for it to finish.The assistant calls the create tool with wait: false and returns the task id. Check on it later with get_task.
Check pricing before creating
Check current Seedream pricing, then create the task if it matches my request.The assistant calls check_pricing and can link to the Seedream model page for the canonical catalog entry.
Configuration
The server resolves auth in this order:
RUNAPI_API_KEYenvironment variable, useful for headless and CI hosts~/.config/runapi/config.json, created by the MCPlogintool orrunapi loginNo key, which still allows
check_pricing
The config file is normally managed by login. A pre-provisioned headless config can use:
{
"apiKey": "your_runapi_key"
}Do not commit real API keys.
Links
Resource | URL |
Seedream model page | |
npm package | |
GitHub repository | |
RunAPI MCP overview | |
RunAPI docs |
License
Licensed under the Apache License, Version 2.0.
Available Tools
6 toolscheck_pricingB
Look up RunAPI pricing for the seedream model line.
Inputs: model, action.
decompose_layersC
Create a Seedream task on RunAPI (decompose layers). Returns a task id, status, and output URLs.
Inputs: image_url (required), size, wait, model, +5 more.
edit_imageC
Create a Seedream task on RunAPI (edit image). Returns a task id, status, and output URLs.
Inputs: source_image_urls (required), seed, wait, model, +11 more.
get_taskA
Fetch the current status and latest result payload for a seedream task.
Inputs: action (required), task_id (required).
loginA
Authenticate RunAPI by opening a browser PKCE login flow and saving the API key to ~/.config/runapi/config.json.
Inputs: force.
text_to_imageB
Create a Seedream task on RunAPI (text to image). Returns a task id, status, and output URLs.
Inputs: seed, wait, model, prompt, +9 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
Each tool has a clearly distinct role: authentication, pricing lookup, task status retrieval, and three separate image-generation modes. The three creation tools are explicitly differentiated by their Seedream task type (decompose layers, edit image, text to image), so an agent should not confuse them.
Most names follow a verb_noun pattern (get_task, check_pricing, decompose_layers, edit_image), but text_to_image is a noun phrase rather than a verb-led action, and login lacks an object. The set remains readable, but the convention is not uniform.
Six tools is well-scoped for a Seedream/RunAPI integration: authentication, pricing, task status, and three generation modes. Each tool has a clear place, with no obvious bloat or thin coverage.
The core lifecycle is covered: login, check pricing, create tasks for all three image operations, and poll task status/results. Minor gaps exist around task management, such as listing tasks, canceling tasks, or logging out/clearing credentials.
Maintenance
Related MCP Connectors
Make images and videos, price requests, list AI models and read your balance, from one API key.
Generate images and videos (GPT Image, Seedance, Kling, Wan, Qwen) with one API key.
130+ AI models for image, video, music, and audio — 18 model families, one RunAPI account.
Generate images and videos with Seedream, GPT Image, Nano Banana, Kling and Seedance on one key.
Related MCP Servers
- AlicenseAqualityBmaintenanceEnables creating and managing GPT Image tasks (edit and text-to-image) via RunAPI, with options to poll status and check pricing.5136 npmApache 2.0
- AlicenseBqualityBmaintenanceEnables AI image and video generation tasks (text-to-image, image-to-video, edit, upscale, etc.) via RunAPI, with support for polling and pricing lookups.1084 npm1Apache 2.0
- AlicenseAqualityBmaintenanceEnables users to create image editing and text-to-image tasks, poll their status, and check pricing for the Nano Banana model line through a RunAPI API key.554 npmApache 2.0
- AlicenseAqualityBmaintenanceEnables creating and managing Runway video generation tasks (text-to-video, video extension) through a single RunAPI key, including task polling and pricing checks.549 npmApache 2.0