all posts in the Tutorial category
maginary has an MCP server. hosted at mcp.maginary.ai, nothing to install: paste one url into Claude Desktop, Cursor, Claude Code or VS Code and your assistant can generate, upscale, vary and animate — and sign up for you in-chat. here's the 30-second setup and what it looks like in practice.
a practical checklist for preparing ai-generated images for portfolios, client work, and social media, covering metadata, captions, borders, and format conversion.
Plain prompts, --flagship, and exact model triggers — the three lanes of Maginary's model selection, and how to get the right model without memorizing a menu.
From "I have an API key" to "I'm rendering the image in my app" in under ten minutes. curl + python + node examples, the full request/response shape, and how to pick between polling and webhooks.
Get a POST when your generation finishes. Payload, headers, HMAC signature verification in Python / Node / Go, retry policy, idempotency — every gotcha we've hit while shipping this in production.
Learn how to create your first AI-generated images and videos with Maginary. From YouTube thumbnails to product shots, we've got you covered.