MCP servers that generate images
Turn prompts into images via Flux, MiniMax, Recraft, and more.
15 servers · Last updated August 5, 2026
TL;DR: These servers wrap text-to-image (and sometimes video/audio) models so your agent can create assets on demand. They differ by underlying model, speed/cost, and whether they also do edits, upscaling, or background removal.
Bottom line: if you only try one, BlenderMCP is the most popular option for this (23,218★). 14 more compared below.
Compare 15 servers
| Server | Transport | Auth | Verified | Stars | Tools for this |
|---|---|---|---|---|---|
| BlenderMCP | Local (stdio) | No auth | — | 23,218 | generate_hyper3d_model_via_text, generate_hyper3d_model_via_images, import_generated_asset +2 |
| MiniMax MCP Server (Official) | Local (stdio) | API key | 1,500 | text_to_image, generate_video | |
| MeiGen AI Design MCP | Local (stdio) | API key | — | 1,457 | generate_image, generate_video |
| figma-mcp-go | Local (stdio) | No auth | — | 1,140 | import_image |
| MCP Mermaid | Local (stdio) | No auth | — | 588 | generate_mermaid_diagram |
| Video Editor MCP server | Local (stdio) | API key | — | 272 | generate-edit-from-videos, generate-edit-from-single-video |
| Comfy Pilot | Local (stdio) | No auth | — | 213 | view_image |
| Aseprite MCP Tools | Local (stdio) | No auth | — | 205 | generate_color_ramp, import_image_as_layer |
| SolidWorks MCP | Local (stdio) | No auth | — | 193 | generate_vba_script |
| Replicate Flux MCP | Local (stdio) | API key | 103 | generate_image, generate_multiple_images, generate_image_variants +1 | |
| openai-gpt-image-mcp | Local (stdio) | API key | — | 102 | create-image, edit-image |
| Recraft MCP Server (Official) | Local (stdio) | API key | 60 | generate_image, image_to_image, vectorize_image | |
| Fal.ai MCP Server | Local (stdio) | API key | 49 | generate_image, generate_image_structured, generate_image_from_image +9 | |
| MCP Image Compression | Local (stdio) | No auth | — | 33 | image_compression |
| Adaptive Cards MCP | Local (stdio) | No auth | — | 27 | generate_card, generate_and_validate |
The servers
Control Blender from Claude and other LLMs for prompt-assisted 3D modeling, scene creation, and asset generation.
Official MiniMax server for TTS, voice cloning, music, image, and video generation.
MCP server for AI image & video generation across 9 models, with a 1,446-prompt library and parallel sub-agent orchestration.
Free, no-rate-limit Figma MCP server with full read/write via a plugin bridge — no REST API token needed.
Generate Mermaid diagrams and charts with AI dynamically, exporting to SVG, PNG, base64, file, or shareable URLs.
Upload, search, edit, and generate videos through Video Jungle from your LLM.
MCP server + embedded terminal that lets Claude Code see, edit, and run your ComfyUI workflows.
Give AI assistants full control over Aseprite to create pixel art and animated sprites.
Automate SolidWorks over COM from your AI client: sketches, features, drawings, exports and VBA generation (alpha).
Community server generating images and SVG assets via Replicate's Flux models.
MCP server for OpenAI GPT-4o / gpt-image-1 image generation and editing.
Official Recraft server for raster and vector image generation, vectorization, and editing.
Community server for image, video, music, and audio generation across fal.ai models.
High-performance MCP server for compressing JPEG, PNG, WebP and AVIF images offline.
Generate, validate, and optimize accessible Adaptive Cards for Teams, Outlook, Copilot, and other Microsoft surfaces.
Use these in a stack
FAQ
Which image-generation MCP is fastest?
Hosted inference providers (fal, Replicate) optimize for latency; the right pick depends on the model you want and your budget per image.
Can these edit existing images, not just generate?
Some expose edit/upscale/variation tools in addition to generation — check each server's tool list on its page.