Roman Numeral Converter

Choose Number → Roman to encode a whole number, or Roman → Number to decode a numeral.

Enter a value and the result updates immediately. Switching direction carries a valid result into the opposite input so it can be converted back.

Each result includes a token breakdown showing the value contributed by every symbol or subtractive pair.

The modern notation used here is built from seven Latin letters: I, V, X, L, C, D, and M.

Symbols normally run from larger to smaller values and are added. For example, MDCLXVI equals 1,000 + 500 + 100 + 50 + 10 + 5 + 1, or 1,666.

Canonical modern notation uses six subtractive pairs: IV, IX, XL, XC, CD, and CM.

Subtraction is local rather than a general rule. Forty-nine is XLIX, not IL, and ninety-nine is XCIX, not IC.

The converter moves through canonical values from largest to smallest: 1,000, 900, 500, 400, 100, 90, 50, 40, 10, 9, 5, 4, and 1.

It appends the largest token that fits, subtracts that value, and continues until nothing remains.

For 944, the decomposition is 900 + 40 + 4, producing CM + XL + IV, or CMXLIV.

The input is normalized to uppercase and checked against the supported letters.

It must then match the selected canonical thousands, hundreds, tens, and ones patterns.

Only after that validation is the numeral decoded, preventing unsupported strings such as IC, VX, or IIV from being assigned a guessed meaning.

Roman numerals were not written in one completely uniform way throughout antiquity, medieval manuscripts, inscriptions, clocks, and decorative uses.

This converter deliberately implements one predictable modern convention so every accepted numeral converts back to exactly the same spelling.

A rejected form is not automatically historically wrong. It means the form falls outside the convention implemented here.

The chart is generated with the same conversion function as the live tool.

The patterns repeat by place value: the ones use I, V, and X, while the tens use X, L, and C.

Some unsupported forms have a common intended value, but noncanonical strings are not always unambiguous.

The converter therefore rejects them instead of silently rewriting the input. The table shows common intentions and the modern form produced for those numbers.

Roman numerals remain common for years on buildings and film credits, monarch and pope numbers, book sections, event editions, clock faces, and outlines.

The year 2026 is MMXXVI.

The converter handles one whole number at a time. A complete date must be split into its day, numeric month, and year before each part is converted.

Clock faces often show IIII for four, while this converter returns IV.

Lowercase input is converted to uppercase before validation.

Unicode compatibility characters such as Ⅻ are normalized with NFKC, so Ⅻ becomes XII and decodes to 12.

For most text, Unicode recommends composing Roman numerals from ordinary Latin letters. The converter therefore copies results as letters such as M, C, X, V, and I.

Using the seven ordinary letters and the selected modern place-value convention gives a clear range from I through MMMCMXCIX.

Larger values have been written with overlines, apostrophus forms, additional symbols, and other historical conventions.

Because those extensions are not one interchangeable plain-text system, the tool stops at 3,999 rather than presenting one extension as universal.

The notation implemented here has no symbol for zero and begins at 1.

Negative numbers, decimal fractions, and historical fraction systems are outside scope.

The converter changes notation only. It does not add, subtract, multiply, or divide Roman numerals.

All conversion and validation run in the browser.

Entered values are not uploaded, added to a public URL, or permanently stored.

Clipboard access is requested only when the copy control is used.