Technology

MB vs. MiB: What Is the Difference?

Compare exact megabyte and mebibyte byte counts, convert both directions, and distinguish storage size from megabits and transfer rates. This guide explains the reasoning, not just the keystrokes.

Published · Updated · 6 min read

Two exact byte quantities

One megabyte, symbol MB, equals exactly 1,000,000 bytes under decimal scaling. One mebibyte, symbol MiB, equals exactly 1,048,576 bytes under IEC binary scaling. MiB is therefore larger than MB even though both names are often informally called “megabytes.”

The ratio is exact: 1 MiB = 1.048576 MB. In the other direction, 1 MB = 0.95367431640625 MiB. Keep the symbols with every result so the number is not detached from its definition.

MB to MiB

Use MiB = MB × 1,000,000 ÷ 1,048,576. Input: 100 MB. Formula: MiB = 100 × 1,000,000 ÷ 1,048,576. Calculation: 95.367431640625. Result: about 95.3674 MiB. Interpretation: both values represent exactly 100,000,000 bytes. Reverse check: 95.367431640625 × 1,048,576 ÷ 1,000,000 = 100 MB.

MiB to MB

Use MB = MiB × 1,048,576 ÷ 1,000,000. Input: 64 MiB. Formula: MB = 64 × 1,048,576 ÷ 1,000,000. Calculation: 67.108864. Result: exactly 67.108864 MB. Interpretation: the decimal number is larger because MB is the smaller unit. Reverse check: 67.108864 × 1,000,000 ÷ 1,048,576 = 64 MiB.

A one-unit definition check

Input: 1 MiB. Formula: MB = 1,048,576 ÷ 1,000,000. Calculation: 1.048576. Result: exactly 1.048576 MB. Interpretation: this is a unit-definition comparison, not a measured file. Reverse check: 1.048576 MB contains 1,048,576 bytes, which is exactly 1 MiB.

Using 1.05 as the factor is a useful two-decimal estimate but not the exact relationship. Across large totals, retain the byte definitions.

Files, memory, and downloads

A file manager, operating system, memory specification, download page, or cloud service may choose different labels and scaling. Read documentation rather than assuming every displayed MB uses the same base. Compression and allocation can also make an on-disk size differ from logical file bytes.

For transfer rates, MB/s is megabytes per second while Mb/s is megabits per second. Converting storage size alone does not predict time; actual throughput and overhead matter.

Capitalization prevents an eightfold error

Uppercase B means byte; lowercase b means bit. 100 MB is 800 Mb before adding a time denominator. MiB should not be written Mib when bytes are intended. The lowercase i in MiB identifies a binary prefix, not a bit.

Ambiguous historical labels such as “MB” used for 1,048,576 bytes should be reported with their source convention. A modern conversion should not silently reinterpret archival data.

Using the Data Storage Converter

The converter accepts MB and MiB as distinct selections and converts through bytes. It also supports larger and smaller decimal and binary byte units. It does not calculate compressed size, allocated disk blocks, memory availability, bandwidth, or download duration.

For exact large byte counts, software may need integer arithmetic because floating-point numbers cannot represent every sufficiently large integer. Display rounding should be separated from the stored byte count.

More MB/MiB examples and implementation checks

Input: 500 MB. Formula: MiB = 500,000,000 ÷ 1,048,576. Calculation: 476.837158203125. Result: about 476.837 MiB. Interpretation: both labels describe exactly 500 million bytes. Reverse check: 476.837158203125 × 1,048,576 = 500,000,000 bytes = 500 MB.

Input: 256 MiB. Formula: MB = 256 × 1,048,576 ÷ 1,000,000. Calculation: 268.435456. Result: exactly 268.435456 MB. Interpretation: powers-of-two memory quantities often produce familiar binary counts but non-round decimal MB values. Reverse check through bytes recovers 256 MiB.

A file shown as 10 MB by one service and roughly 9.54 MiB by another may be the same byte sequence. Before diagnosing a mismatch, compare the exact bytes and the label convention. If byte totals differ too, investigate compression, metadata, allocated size, transfer packaging, or a genuinely different file. Unit conversion alone cannot select among those explanations.

Rounding can be material near upload limits. A service advertising a 100 MB limit may mean exactly 100,000,000 bytes, while a program showing MiB may display that as about 95.37 MiB. Do not assume an ambiguous “100 MB” means 100 MiB. The service's documented byte limit controls, and a general converter cannot test policy enforcement or upload overhead.

Implementation should convert both units through integer bytes and attach the symbol to the result. Test exact anchors of 1 MB, 1 MiB, 100 MB, and 64 MiB plus zero and large values. Do not lowercase B during UI normalization. If a source uses MB ambiguously for a binary quantity, keep that provenance in text rather than silently changing the stored definition.

Input: 1,000 MiB. Formula: MB = 1,000 × 1,048,576 ÷ 1,000,000. Calculation: 1,048.576. Result: exactly 1,048.576 MB. Interpretation: the same bytes exceed one decimal gigabyte but remain below one binary gibibyte. Reverse check: 1,048.576 MB × 1,000,000 ÷ 1,048,576 = 1,000 MiB.

For upload, memory, and software requirements, the documented unit controls. A program requiring 512 MiB is asking for 536.870912 MB, while a 512 MB limit is only about 488.28125 MiB. Treating those numbers as identical can put a value on the wrong side of a threshold. Confirm whether the system evaluates exact bytes or a rounded display.

Compression changes byte count; unit conversion does not. A 200 MB archive extracted into 800 MiB of files has changed content representation as well as unit. Similarly, an on-disk allocation can exceed a logical file size because storage is assigned in blocks. Compare the same size concept before attributing a difference to MB versus MiB.

For communication, show the exact byte count when ambiguity matters and use one standardized symbol consistently. A sentence such as “100,000,000 bytes (100 MB, about 95.37 MiB)” is reproducible. Writing only “100 meg” is not. The converter can generate the alternate value but cannot repair an undefined source label.

A simple magnitude check is that a fixed byte count has a larger number in MB than in MiB because MB is the smaller unit. If 100 MB converts to more than 100 MiB, the ratio was inverted. This check does not resolve an ambiguous source label, so inspect documentation before arithmetic.

Zero MB equals zero MiB. Negative file size or capacity is ordinarily invalid, though a signed change in usage can be negative. Interfaces should distinguish a quantity from a delta and validate accordingly. The converter performs numeric scaling and cannot infer which meaning the user intended.

For very large values, convert through bytes with integer-safe arithmetic and round only the user-facing unit. A rounded MiB total should not be used as the authoritative basis for billing, quotas, offsets, or file integrity. Preserve exact byte counts wherever those decisions depend on individual bytes.

When copying results into prose, avoid claiming that one convention is inherently the “real” size. The byte count is the common quantity; MB and MiB are two defined ways to scale it. Criticism should focus on ambiguous or incorrect labeling, not on the existence of either unit. Showing bytes plus the chosen symbol gives readers enough information to reproduce the result.

The same ratio applies to fractional values. Preserve enough internal precision that converting a small MB quantity to MiB does not round to zero prematurely, then select display decimals appropriate to the user and scale.

Frequently asked questions

How many bytes are in one MB?

Exactly 1,000,000 bytes.

How many bytes are in one MiB?

Exactly 1,048,576 bytes.

How many MB are in one MiB?

Exactly 1.048576 MB.

Is MB the same as Mb?

No. MB is megabytes; Mb is megabits and is eight times smaller for the same prefix count.

Why does software show a different file-size number?

It may use different decimal or binary scaling, rounding, compression, or allocated-size reporting.

Sources

References used to support definitions and interpretation in this guide.

  1. IEC 80000-13:2025 — Quantities and units, Part 13: Information science and technology
  2. BIPM — The International System of Units (SI Brochure), 9th edition