Add Text to Image

Adding text to an image draws words onto the picture itself, so they travel with the file. The caption is part of the pixels from then on, which is what you want for a social post, a screenshot you are annotating, a meme, a quote card or a simple watermark. It is different from a caption typed into a post, which lives beside the image and disappears when someone saves or reposts it.

The part that looks easy and is not is making the text readable over a photograph. A photo is not a flat colour: white text is invisible over a bright sky and black text disappears into a shadow, often inside the same caption. That is why this tool offers an outline and a drop shadow rather than only a colour. An outline in the opposite tone of the text is the reliable fix, because it gives every letter its own contrast regardless of what is behind it, which is exactly how subtitles have always been done.

Sizing is the other thing worth getting right. Picking a font size in pixels means something different on a 600 pixel image than on a 4000 pixel one, so the same number gives a caption you cannot read on one and one that fills the frame on the other. Auto-fit takes the opposite approach: you say how much of the image width the text should occupy, and the size is solved for that, by binary search over the wrapped layout. Resize the box or edit the words and the size re-solves, so a caption keeps its proportions on any photo.

Wrapping follows two rules that come from reading rather than from arithmetic. A line break you type is kept as a line break, because someone writing two lines means two lines. A single word wider than the box is left whole instead of being split mid-word, since "unbeliev" above "able" reads worse than a slight overflow, and auto-fit resolves it by shrinking the text anyway.

Nine anchor positions cover where a caption usually goes: the three across the top, the three across the middle, and the three along the bottom, each inset by a padding you control. Anchoring rather than dragging to coordinates means the caption stays where you put it relative to the frame, so a bottom-centre caption is still bottom-centre after you change the font size or add a line.

The image is decoded, drawn and exported in your browser. Nothing is uploaded, so a screenshot with a customer name in it, or a photo you have not published yet, does not pass through a server on the way to having a caption.