For hero or article images, use an image model included with your plan if one is available. For images billed to your own provider account, PixelClaw supports OpenAI (GPT Image) and Google Gemini (Nano Banana image models). Your own OpenRouter, Claude, DeepSeek, MiniMax, and Kimi connections are for writing in PixelClaw; they do not generate images here.
A Blog Site owner or admin can open Settings → Integrations → Writing model and billing → Included with your plan → Images to see the available plan models and price per image. When you request a standalone image, check its model and quote. An AI-written draft can also make a hero image automatically when the feature is available. Rewriting a draft can keep its existing picture.