Image File Size Estimator

Estimate image file size by format and quality (JPEG, PNG, WebP, AVIF) from width and height, using typical bits-per-pixel figures.

Result

Loading calculator…

—

How to use this calculator

Enter the image width and height in pixels.
Choose the file format (JPEG, WebP, AVIF, PNG, or uncompressed BMP).
For lossy formats, pick a quality level.
Read the estimated file size and the assumed bits-per-pixel; expect ±50% versus real files.

About this calculator

How big will an image file be? It depends on the pixel dimensions, the file format, and — for lossy formats — the quality setting. This tool estimates file size from a typical bits-per-pixel figure for photographic content: it multiplies the pixel count by the bits each pixel consumes on average, then converts to bytes. JPEG, WebP, and AVIF are lossy formats that trade detail for size, with AVIF the most efficient and JPEG the least; WebP typically saves about 25–35% over JPEG and AVIF around 50%, at comparable quality. PNG is lossless and best for graphics, screenshots, and images with sharp edges or transparency, but it produces large files for detailed photos. BMP is fully uncompressed. Because real file size depends heavily on image content, treat these numbers as a typical-photo ballpark rather than an exact figure.

How it works — the formula

Bytes ≈ Width × Height × bits-per-pixel ÷ 8
Uncompressed (24-bit) = Width × Height × 3 bytes

File size scales with pixel count and the per-pixel bit budget the format/quality implies. Lossless and uncompressed formats use fixed high per-pixel costs; lossy formats use much lower, quality-dependent figures.

Source: Google — WebP compression study (format size comparison)

Worked examples

Example 1

1920×1080 JPEG, medium (~1.0 bpp)

Inputs:: width=1920, height=1080, format=jpeg, quality=medium
Output:: ≈ 253 KiB

Example 2

4000×3000 JPEG, high (~2.0 bpp)

Inputs:: width=4000, height=3000, format=jpeg, quality=high
Output:: ≈ 2.86 MiB

Example 3

1920×1080 AVIF, medium (~0.5 bpp)

Inputs:: width=1920, height=1080, format=avif, quality=medium
Output:: ≈ 127 KiB

Limitations

Bits-per-pixel figures assume photographic content; graphics differ greatly.
PNG size is highly content-dependent — the estimate is a rough upper-mid guess.
Does not model metadata (EXIF), thumbnails, or alpha channels precisely.

A planning estimate (±50%). For exact sizes, encode the actual image — see the JPEG quality optimizer tool.

Frequently asked

How is image file size estimated?+

By multiplying the number of pixels (width × height) by a typical bits-per-pixel figure for the chosen format and quality, then dividing by 8 to get bytes. For example, a 1920×1080 JPEG at medium quality (~1 bit/pixel) is about 2,073,600 ÷ 8 ≈ 253 KiB.

Why do JPEG, WebP, and AVIF differ in size?+

They use progressively more advanced compression. For similar visual quality, WebP is typically 25–35% smaller than JPEG, and AVIF roughly 50% smaller, because their encoders model image data more efficiently. The trade-off is encoder/decoder support and speed.

Why is PNG so much larger for photos?+

PNG is lossless — it preserves every pixel exactly, which is great for graphics and text but inefficient for the continuous tones and noise of photographs. A photographic PNG can be several times larger than a high-quality JPEG of the same image.

How accurate is this estimate?+

It is a ballpark for typical photographic content, easily off by ±50% or more. Real size depends on the actual detail: a flat blue sky compresses to almost nothing, while busy foliage or grain inflates the file. Use it for planning, not exact budgeting.

When should I use each format?+

Use JPEG for broad compatibility with photos, WebP or AVIF for the web when you want smaller files and your audience's browsers support them, and PNG for screenshots, logos, line art, or anything needing transparency or pixel-perfect edges.

What does the uncompressed reference mean?+

It is the size the image would be with no compression at 24 bits per pixel (8 bits each for red, green, blue). Comparing it to the estimate shows the compression ratio you are getting — often 10–50× smaller for JPEG.