Type lab
A typeface is not chosen by its name
Testing starts with the characters people actually write: č, ć, đ, š, ž and localised Serbian Cyrillic italics. Character, rhythm and loading come next.
What the lab checks
I checked 103 Google Fonts with Cyrillic for Serbian character coverage and localised italic forms. Twenty-three contained full Serbian italics; another thirteen use a neutral slanted form without introducing Russian shapes.
Four tests before selection
- Check č, ć, đ, š and ž in every required weight and style.
- Check Serbian Cyrillic characters ђ, ј, љ, њ, ћ and џ.
- Inspect б, г, д, п and т in italics when the language is set to Serbian.
- Measure the font subsets the page actually downloads.
Why language metadata matters
A capable font contains both Russian and Serbian italic glyphs. The browser selects the local forms through the page language. A Latin-script Serbian page can still activate the right Cyrillic forms because the language remains Serbian.
The result is more than style
A wrong font can insert a fallback face halfway through a word, clip a diacritic during animation or change the form of a Cyrillic letter. A good choice protects both the brand voice and the technical whole.
The full 103-font filter is currently presented in Serbian on the tested Cyrillic font catalogue.