Compress a JPG to 100KB

A hundred kilobytes is where JPG stops fighting you. The format has enough room to describe smooth gradients without banding and enough left over for texture, which is why photographs at ordinary web dimensions look clean here in a way they do not at half the budget.

How should the settings be chosen?

    Comfortable territory for the format

    At fourteen hundred pixels wide, a typical photograph reaches this figure at a quality setting in the low eighties, which sits above where JPG artefacts become apparent to a casual viewer. Skies stay smooth rather than stepping into bands, and skin tones keep their gradation.

    That means you can often leave the dimensions alone. Where a fifty kilobyte target forces a resize, this one frequently does not, which preserves detail that a downscale would have removed permanently.

    Deciding between width and quality

    If the result lands under target with quality still high, raise the maximum width and re-run — you have headroom you are not using. If it only just fits at a low quality, do the opposite. The comparison slider settles the question faster than reasoning about it, and it is the reason the slider is on every result.

    Keeping JPG when you could use WebP

    The honest position is that WebP would produce a better picture at this size. Stay with JPG when the destination requires it, when the file will pass through systems you do not control, or when someone will open it in software of unknown age. Otherwise the newer format is a free improvement.

    Questions

    Is 100KB enough for a full width photograph on a web page?
    For most pages, yes, particularly at around fourteen hundred pixels wide. A very wide hero on a large display may justify more, and a photograph with heavy texture such as foliage benefits from two hundred. Compare the results before assuming you need the larger file.
    Why does one photo hit 100KB easily and another struggle?
    Because encoders spend their budget on detail. A portrait against a plain wall is largely smooth and compresses cheaply, while a photograph of a crowd or a forest changes in every direction at once and is expensive everywhere. Two files of identical dimensions can differ several times over in size for that reason.
    Should I sharpen the image before compressing?
    Generally not. Sharpening adds high contrast edge detail, which is exactly what a lossy encoder finds expensive, so a sharpened image reaches a given size at lower quality than the same image untouched. If you want sharpening, apply it after compression, in whatever displays the file.
    Does the age of the JPG format mean I should avoid it?
    No. It remains the most widely accepted photographic format anywhere, opens in software decades old, and at a comfortable budget produces results very close to newer formats. Its weaknesses appear when you squeeze it hard or feed it sharp edged graphics. Choose it deliberately for compatibility, and choose something newer when nothing forces your hand.

    Related