PixelMCP is an image studio that runs inside your AI assistant. Connect it once over the Model Context Protocol (MCP), then ask for images in plain language and get finished files back in the same chat.
Models: Nano Banana 2 and Pro, GPT Image 2, FLUX.2 Pro and Klein, Seedream 5.0 Pro and Recraft V4.1, all on one credit balance.
Works with Claude, ChatGPT, Gemini, Cursor, VS Code, Claude Code, Codex and any other MCP client.
Ask for up to four variations at once or try the same brief on different models, save a brand kit so every image stays on-brand, and keep every generation in your library. Pay per image with credits that never expire. New accounts get 100 free credits.
What you can make: social posts in square, portrait and story sizes; studio and lifestyle product shots for your store and ads; logo directions you can refine; hero images, banners and announcement assets for a launch; many ad variations with different hooks and angles; blog headers and social cards straight from your editor; and consistent spot illustrations and app icons in your palette.
How it works: add PixelMCP to your assistant in about a minute, save your colors, fonts and tone once as a brand kit, ask in plain language, then pick the best take, ask for tweaks and download the full-resolution files. You never leave the conversation.
Pricing is simple: buy credit packs and spend them per generation, with no subscription or seat fee. Bigger packs make every image up to 17% cheaper. Video and audio models such as Veo, Kling and Seedance are coming soon on the same credits.
