kie-ai-mcp-server
Universal MCP server for all Kie AI image generation and editing models, enabling text-to-image, image-to-image, and image editing via natural language.
README
Kie AI MCP Server
Universal MCP server for all Kie AI image generation and editing models. One server, every model — pick the one you want at call time.
What's New in v2.1
- Seedream 5 Lite Text-to-Image (
seedream/5-lite-text-to-image) — ByteDance's latest lightweight generation model with high-quality output by default. - Seedream 5 Lite Image-to-Image (
seedream/5-lite-image-to-image) — Transform existing images with natural language. Change lighting, styles, effects, and more. Defaults to high quality output.
Supported Models
Generation (text-to-image / image-to-image)
| Model ID | Name | Highlights |
|---|---|---|
nano-banana-pro |
Nano Banana Pro | Best all-around. 4K, text rendering, up to 8 ref images |
google/nano-banana |
Nano Banana | Fast, affordable Gemini-based generation |
nano-banana-2 |
Nano Banana 2 | Second-gen improvements |
seedream/5-lite-text-to-image |
Seedream 5 Lite | ByteDance's latest. High quality by default |
seedream/5-lite-image-to-image |
Seedream 5 Lite I2I | Transform images with natural language. High quality default |
seedream/4.5-text-to-image |
Seedream 4.5 | ByteDance. Photorealistic, 2K/4K quality modes |
flux-2/pro-text-to-image |
Flux 2 Pro | Photoreal detail, accurate text rendering |
flux-2/pro-image-to-image |
Flux 2 Pro I2I | Style transfer, outfit changes, up to 8 refs |
flux-2/flex-text-to-image |
Flux 2 Flex | Good quality/cost ratio |
flux-2/flex-image-to-image |
Flux 2 Flex I2I | Flexible transformations, lower cost |
flux-1/kontext |
Flux 1 Kontext | Context-aware generation |
imagen-4 |
Imagen 4 | Google's latest, excellent fidelity |
ideogram/v3 |
Ideogram V3 | Best-in-class text/typography in images |
ideogram/character |
Ideogram Character | Consistent character generation |
4o-image |
4o Image | OpenAI 4o-based generation |
z-image |
Z-Image | Alternative generation model |
Editing (image-edit)
| Model ID | Name | Highlights |
|---|---|---|
seedream/4.5-edit |
Seedream 4.5 Edit | Material, lighting, style changes. 4K output |
google/nano-banana-edit |
Nano Banana Edit | Gemini-based editing. Fast and cheap |
qwen/image-edit |
Qwen Image Edit | Alibaba's general-purpose editor |
Tools
| Tool | Description |
|---|---|
kie_list_models |
List all available models with capabilities and parameters |
kie_generate_image |
Generate an image (text-to-image or image-to-image) |
kie_edit_image |
Edit an existing image with natural language |
kie_get_task_status |
Check task status and get result URLs |
Quick Start (npx — no install needed)
1. Get your API key
Go to kie.ai/api-key and create an API key.
2. Add to Claude Desktop / Claude Code
Add this to your MCP settings (e.g. claude_desktop_config.json):
{
"mcpServers": {
"kie-ai": {
"command": "npx",
"args": ["-y", "kie-ai-mcp-server"],
"env": {
"KIE_AI_API_KEY": "your_api_key_here"
}
}
}
}
That's it. npx downloads and runs the server automatically — no cloning, no building.
Alternative: Install from source
git clone https://github.com/spiritbird-ai/kie-ai-mcp-server.git
cd kie-ai-mcp-server
npm install
npm run build
Then use this config instead:
{
"mcpServers": {
"kie-ai": {
"command": "node",
"args": ["/path/to/kie-ai-mcp-server/dist/index.js"],
"env": {
"KIE_AI_API_KEY": "your_api_key_here"
}
}
}
}
Usage Examples
Generate an image
Use kie_generate_image with model "nano-banana-pro", prompt "A cozy coffee shop interior with warm lighting, watercolor style", aspect_ratio "16:9", resolution "2K"
Edit an image
Use kie_edit_image with model "seedream/4.5-edit", prompt "Change the background to a tropical beach at sunset", image_urls ["https://example.com/my-photo.jpg"]
Check task status
Use kie_get_task_status with task_id "abc123..."
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
KIE_AI_API_KEY |
Yes | — | Your Kie AI API key |
KIE_AI_BASE_URL |
No | https://api.kie.ai/api/v1 |
API base URL |
KIE_AI_TIMEOUT |
No | 60000 |
Request timeout in ms |
How It Works
All Kie AI models share the same API pattern:
- Create task:
POST /jobs/createTaskwith{ model, input } - Poll status:
GET /jobs/recordInfo?taskId=... - Get result: When
state === "success", the result URLs are inresultJson
This server wraps that pattern into clean MCP tools with model-specific parameter validation.
Adding New Models
To add a new Kie AI model, simply add an entry to the MODEL_REGISTRY array in src/models.ts. No other changes needed — the tools dynamically adapt to the registry.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.