Basic Multilingual Plane (BMP)

What is the Basic Multilingual Plane?

The Basic Multilingual Plane (BMP) is Plane 0 of the Unicode code space, covering code points U+0000 through U+FFFF — a range of exactly 65,536 positions. It was designed to hold all the characters needed for modern text in the world's actively used scripts, and it largely succeeded: the Latin, Greek, Cyrillic, Arabic, Hebrew, Devanagari, CJK, and dozens of other scripts all fit within the BMP.

The BMP's boundaries matter beyond just organization. Because BMP code points fit in a single 16-bit value (0x0000–0xFFFF), they can be stored as a single code unit in UTF-16, and they are the direct ancestors of UCS-2, the predecessor to UTF-16.

What Lives in the BMP

The BMP is organized into blocks — contiguous ranges assigned to specific scripts or purposes. Notable regions include:

The Surrogate Hole

One important quirk: the range U+D800–U+DFFF (2,048 code points) is permanently reserved for surrogates — the mechanism UTF-16 uses to encode characters above U+FFFF. These code points can never be assigned to real characters. You will sometimes see UTF-16 described as covering the "BMP minus surrogates."

BMP vs Supplementary Characters

Any character with a code point above U+FFFF is a supplementary character and requires special handling in encodings optimized for the BMP:

In UTF-16, supplementary characters require a surrogate pair — two 16-bit code units working together. Most emoji fall into Plane 1 (U+1F000+) and are therefore supplementary.

Historical Significance

Early Unicode architects hoped that 65,536 code points would be enough for all world languages forever. They were wrong. By Unicode 2.0, it was clear that CJK ideographs alone would eventually overflow the BMP, and the standard was extended to 17 planes. This is why legacy systems built on UCS-2 (a fixed-width 16-bit encoding) failed: they could only represent BMP characters.

Common Pitfalls

UCS-2 vs UTF-16: UCS-2 encodes only the BMP using fixed 2-byte units. UTF-16 extends UCS-2 with surrogate pairs for supplementary characters. Many old systems claiming "Unicode support" actually only support UCS-2 (BMP-only).

Emoji in JavaScript: Because JavaScript strings are UTF-16, emoji (Plane 1) have .length of 2, not 1. Iterating with spread or Array.from() corrects this.

"🎉".length      // 2 (two UTF-16 code units)
[..."🎉"].length // 1 (one Unicode code point)

Quick Facts