AI Create
Images, logos, banners, social posts
- PDF to ImageRender one page of a PDF as a PNG image. Use when the user needs a PDF page as an image (for a thumbnail, preview, or to paste elsewhere).pure
- OCRExtract text from an image (a photo of a document, a screenshot, a scanned page). Use when the user has an image and wants the text out of it.ai
- Background RemovalRemove the background from an image, leaving a transparent PNG. Use when the user wants a subject cut out of its background. Requires REMOVE_BG_API_KEY.pure
- UpscalingIncrease an image's resolution using high-quality resampling. Use when the user needs a bigger version of an image. This is Lanczos resampling, not ML super-resolution — it sharpens scaling artifacts, it does not invent detail that isn't in the source.pure
- Logo CreationGenerate a simple wordmark-style logo (brand name in a colored circle) from a brand name and style. Use when the user wants a quick logo mark. This is a template renderer, not a diffusion model — it produces a clean typographic mark, not illustrated artwork.ai
- Banner MakerDraft banner copy and render it as a web banner graphic (e.g. for a sale, launch, or announcement). Use when the user wants a banner image with a headline. Template-rendered, not a diffusion model.ai
- Social Media PostsDraft a caption, hashtags, and a matching graphic for a social media post. Use when the user wants a ready-to-post social update. Template-rendered graphic, not a diffusion model.ai
- Thumbnail CreatorDraft a punchy, clickable title and render a video thumbnail graphic (defaults to 16:9 for YouTube). Use when the user wants a thumbnail for a video. Template-rendered, not a diffusion model.ai
- Image GenerationGenerate an original image from a text prompt. Use when the user wants artwork, a photo, or an illustration created from a description. Claude itself has no image model, so this calls OpenAI's image API — requires OPENAI_API_KEY.pure