`search_models`

Pack: civitai · Endpoint: https://gateway.pipeworx.io/civitai/mcp

Search Civitai’s catalog of community AI image-generation models (Stable Diffusion checkpoints, LoRAs, textual-inversion embeddings, ControlNets) by text query and/or type. Returns download stats, base model (SDXL / Pony / Flux / SD 1.5), creator and trained trigger words. Defaults to Safe-For-Work (mature content excluded). Keyless.

Parameters

Name	Type	Required	Description
`query`	string	no	Text search over model names/descriptions, e.g. “realistic”, “anime”, “pixel art”.
`type`	string	no	Model type filter. One of: Checkpoint, LORA, TextualInversion, LoCon, Controlnet.
`sort`	string	no	Sort order: “Highest Rated”, “Most Downloaded”, or “Newest”. Default “Most Downloaded”.
`include_mature`	boolean	no	Include mature/NSFW models. Default false (Safe-For-Work — sends nsfw=false to the API).
`limit`	number	no	Number of results, 1-25. Default 10.

Example call

Arguments

{
  "query": "realistic portrait",
  "type": "Checkpoint",
  "sort": "Most Downloaded",
  "limit": 10
}

curl

curl -X POST https://gateway.pipeworx.io/civitai/mcp \
  -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"search_models","arguments":{"query":"realistic portrait","type":"Checkpoint","sort":"Most Downloaded","limit":10}}}'

TypeScript (@pipeworx/sdk)

import { Pipeworx } from '@pipeworx/sdk';
const pipeworx = new Pipeworx();

const result = await pipeworx.call('search_models', {
  "query": "realistic portrait",
  "type": "Checkpoint",
  "sort": "Most Downloaded",
  "limit": 10
});

More examples

{
  "query": "anime",
  "type": "LORA",
  "include_mature": false,
  "limit": 5
}

Connect

Add this to your MCP client config, or use one-click install buttons:

{
  "mcpServers": {
    "civitai": {
      "url": "https://gateway.pipeworx.io/civitai/mcp"
    }
  }
}

See Getting Started for client-specific install steps.

search_models