Turkish
Links on this page to features marked Preview describe designs that are not available yet.
How to get natural Turkish out of Syntic models, and what to check before you ship. The general rules for every language are on Work in other languages.
At a glance
| Language tags | tr-TR |
| Script and direction | Latin script with ç ğ ı İ ö ş ü, left to right |
| Token use | Measure with token counting on your own text; non-Latin scripts and long compounds usually take more tokens per sentence than English. |
Register and form of address
Siz is the polite form for customers; sen is informal. Formal written Turkish often uses -iniz/-ınız endings throughout.
Numbers, dates and currency
Decimal comma, dot for thousands (1.234,50 ₺). Dates: 5 Kasım 2026, 05.11.2026.
Common pitfalls
- Dotted and dotless i: lower-casing I gives ı and upper-casing i gives İ in Turkish. Code that changes case with English rules corrupts words; use locale-aware functions.
- Agglutination makes words long and suffixes carry meaning; placeholders like
{name}need correct vowel-harmony suffixes (Ali’ye, Ayşe’ye). Prefer phrasing that avoids suffixing inserted names.
A system prompt line that works
Writing the language instruction in Turkish itself is a strong signal:
Türkçe yanıt ver, kullanıcıya "siz" diye hitap et, nazik ve kısa ol.(Answer in Turkish, address the user with siz, be polite and brief.)
Keep the rest of your instructions in whichever language your team maintains best; the model follows mixed-language system prompts reliably, and the reply language follows this line.
Test before launch
- Collect 30 or more real Turkish messages, including short, misspelt and mixed-language ones.
- Have a fluent speaker grade replies for correctness, register and naturalness; then calibrate a rubric grader on their grades (Evaluate).
- Check rendering in your actual UI: line breaks, truncation and fonts.
- Re-run the set whenever you change the prompt or the model.
Was this page helpful?