civitai-mcp-ultimate

The most complete MCP server for Civitai API. Search models, browse top images with prompts, download LoRAs/Checkpoints, analyze trends — all from your AI assistant.

All API parameters verified against real Civitai API on 2026-03-24. No guessing, no copying from broken implementations.

Features

Meilisearch-powered search — fast, accurate text search (Civitai REST API search is broken since May 2025)
Search models — Checkpoints, LoRAs, ControlNets with 40+ base model filters (SD 1.5 → Flux.2)
Top images with prompts — sort by reactions, comments, collections. Get full generation params
Download commands — curl/PowerShell commands with auth, ComfyUI path auto-mapping
NSFW support — full NSFW access with API key (None/Soft/Mature/X filtering)
Image/video cache — auto-download previews to ~/.civitai-mcp-cache/, 24h auto-cleanup
Bilingual output — English and Russian (CIVITAI_MCP_LANG=ru)
Async & fast — httpx async client, retry with backoff on rate limits
14 tools covering 100% of the Civitai public REST API v1

Quick Start

Install from PyPI

pip install civitai-mcp-ultimate

Claude Code

claude mcp add civitai -e CIVITAI_API_KEY=your_key_here -- uvx civitai-mcp-ultimate

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "civitai": {
      "command": "uvx",
      "args": ["civitai-mcp-ultimate"],
      "env": {
        "CIVITAI_API_KEY": "your_key_here"
      }
    }
  }
}

pip

pip install civitai-mcp-ultimate
civitai-mcp-ultimate

Tools

Models (6 tools)

Tool	Description
`search_models`	Search via Meilisearch (text) or REST API (filters). Type, base model, tag, creator, sort
`get_model`	Full model details by ID
`get_model_version`	Version details: files, trigger words, download URLs
`get_model_version_by_hash`	Find model by file hash
`get_top_checkpoints`	Top checkpoints by base model (SDXL, Flux, Pony...)
`get_top_loras`	Top LoRAs by base model

Images (4 tools)

Tool	Description
`browse_images`	Browse images with filters and sorting
`get_top_images`	Top images by reactions/comments/collections
`get_model_images`	Example images for a model with prompts
`get_image_generation_data`	Full generation params from top images

Creators & Tags (2 tools)

Tool	Description
`get_creators`	Browse/search model creators
`get_tags`	Browse/search model tags

Downloads (2 tools)

Tool	Description
`get_download_url`	Authenticated download URL
`get_download_info`	Download commands (curl/PowerShell) + ComfyUI path mapping

Supported Filters

Base Models (all verified working)

SD 1.5, SD 1.5 LCM, SD 1.5 Hyper, SD 2.0, SD 2.1, SDXL 1.0, SDXL Lightning, SDXL Hyper, Flux.1 S, Flux.1 D, Flux.1 Krea, Flux.1 Kontext, Flux.2 D, Flux.2 Klein 9B/4B, Pony, Pony V7, Illustrious, NoobAI, Anima, ZImageBase, AuraFlow, Chroma, HiDream, Hunyuan 1, Kolors, PixArt a/E, Lumina, CogVideoX, Hunyuan Video, LTXV/LTXV2, Mochi, Wan Video (all variants)

Model Types

Checkpoint, LORA, LoCon, TextualInversion, Hypernetwork, Controlnet, Poses, AestheticGradient, Wildcards, MotionModule, VAE, Upscaler, Workflows, Other

Image Sort

Most Reactions, Most Comments, Most Collected, Newest, Oldest

Configuration

Variable	Required	Default	Description
`CIVITAI_API_KEY`	Recommended	—	API key for NSFW access + higher rate limits
`CIVITAI_MCP_LANG`	No	`en`	Output language: `en` or `ru`
`COMFYUI_MODELS_PATH`	No	—	ComfyUI models path for download commands
`MEILISEARCH_KEY`	No	built-in	Meilisearch search-only key (public, has default)

Get your API key: Civitai Account Settings

Examples

> Search for the top Flux LoRAs this month
> Show me the most popular SDXL checkpoints
> Get the best prompts for model 12345
> Give me download commands for this LoRA into my ComfyUI
> Find NSFW LoRAs for Illustrious sorted by reactions

Development

git clone https://github.com/timoncool/civitai-mcp-ultimate
cd civitai-mcp-ultimate
pip install -e ".[dev]"
pytest

Автор

Nerual Dreming (t.me/nerual_dreming) — neuro-cartel.com | основатель ArtGeneration.me

Другие open-source проекты

Проект	Описание
Foundation Music Lab	Генерация музыки + таймлайн-редактор
VibeVoice ASR	Распознавание речи (ASR)
LavaSR	Портативный ASR на базе Faster-Whisper
Bulka	Браузерный синтезатор live-coded музыки
mock.dog	AI-генератор мокапов для маркетплейсов

License

MIT

Если проект полезен — поставьте звездочку!

README.ru.md