Why is byte count different from character count?

UTF-8 encodes ASCII letters in 1 byte, but European accents take 2, most Asian scripts 3, and emoji 4 bytes each — so 'café 😀' is 6 characters but 10 bytes. Systems with size limits count bytes, not characters.

What does this matter for databases?

Column limits like VARCHAR(255) in MySQL's utf8mb4 measure characters but index limits measure bytes; other systems cap raw bytes. Text that fits by character count can still overflow byte limits when emoji or non-Latin scripts are involved.

How do bytes affect SMS length?

GSM-encoded SMS holds 160 characters, but one emoji forces UCS-2 encoding, dropping the limit to 70 — the counter shows which encoding your message triggers and how many segments it will consume.

What is the difference between UTF-8 and UTF-16 sizes?

UTF-8 favors Latin text (1 byte per ASCII char); UTF-16 uses 2 bytes minimum but handles Asian scripts in 2 bytes where UTF-8 needs 3. The counter reports both so you can compare storage costs.

Is the byte counter accurate for emoji with modifiers?

Yes — skin-tone modifiers and ZWJ sequences (family emoji) are counted at their true byte weight, which can reach 25+ bytes for a single visible glyph.

Byte Counter — Count Bytes, UTF-8 Size of Text Online, Free...

Remove Ads

Share on Social Media:

Byte Counter — Because Systems Count Bytes, Not Characters

Your string is 40 characters and the API rejects it for exceeding a 64-byte limit — welcome to Unicode. This free counter shows the exact UTF-8 and UTF-16 byte sizes of any text alongside its character count, exposes which characters inflate the total, and estimates SMS segmentation. Instant, in-browser, no signup.

Byte Weights by Character Type (UTF-8)

Characters	Bytes each	Example
ASCII (a–z, 0–9, punctuation)	1	hello = 5 bytes
Latin accents, Cyrillic, Greek	2	café = 5 bytes
Chinese, Japanese, Korean, Bengali	3	东京 = 6 bytes
Emoji, rare symbols	4	😀 = 4 bytes
ZWJ emoji sequences	11–28	👨‍👩‍👧 = 18 bytes

Where Byte Limits Bite

SMS — one emoji flips the whole message from 160-char GSM to 70-char UCS-2 encoding; multi-segment messages multiply cost. The counter predicts segments before you send.
Databases — index key limits (e.g. 767 or 3072 bytes in MySQL) are byte-measured; a 255-character utf8mb4 column can occupy 1,020 bytes.
APIs and headers — payload and header caps, push-notification limits (APNs 4 KB), and metadata fields all measure bytes.
Filenames and URLs — 255-byte filesystem limits shrink fast with non-Latin names.

UTF-8 vs UTF-16: Storage Trade-Off

English-heavy content is ~half the size in UTF-8; CJK-heavy content is smaller in UTF-16 (2 vs 3 bytes per character). Seeing both figures side by side makes encoding decisions for storage and transfer concrete instead of theoretical.

Free and Instant

All counting is local — paste anything without it leaving your device. Pair with our character counter for platform-limit checks.

Byte Counter

Result