Languages & cost tracking
This page covers two things that happen automatically: replying in the visitor's language, and tracking what your bot costs.
Multilingual replies
Promptly detects the visitor's language and replies in it — no setup required. The bot reads the first part of each visitor's first message to decide the language, then locks it in for the rest of that conversation so it never switches mid-chat. It also matches the right script (for example Latin vs. Cyrillic).
Detection handles common European languages including English, German, French, Italian, Spanish, Turkish, and Serbian (both Latin and Cyrillic). Very short or emoji-only messages ("ok", "👍") fall back to your account's default language instead of guessing.
The underlying search uses a multilingual embedding model (bge-m3), so a visitor can ask in one language about content you wrote in another and still get relevant results. See How retrieval works.
🖼️ [Image] — The same bot answering the same question in three different languages.
You can set a default language for your account, used for canned messages and as the fallback when detection is uncertain. You don't need to write language instructions into your system prompt.
Cost tracking
Every AI message consumes tokens, and those tokens are billed on your own provider account — Promptly doesn't resell tokens or meter your conversations, so there's no message allowance to run out of and no per-message charge.
- Who bills you for tokens — your AI provider (OpenAI, Anthropic, Gemini, GLM, or Kimi), directly, at their list price. Your Promptly subscription pays for the modules and the infrastructure we run for you: knowledge search, storage, and crawling.
- What Promptly does meter — the caps that come with your subscription, all tied to that infrastructure: knowledge base documents, content size, and crawl volume. See Pricing & limits.
- If your key runs out of credit — the assistant surfaces it right away instead of failing silently, and you can add a backup provider key for automatic failover.
🖼️ [Image] — A usage card showing knowledge base usage against your subscription's limits.
Keeping costs down
- Use
gpt-4o-miniand a modest Max Tokens for most bots — see Model & behavior settings. - Keep your system prompt short; it's billed on every message — see Writing a system prompt.
- Lower Max History if visitors rarely reference earlier turns.
Detailed per-conversation figures live in your analytics. The dashboard reports token usage so you can see what drives your cost, but the authoritative bill for tokens is the one your AI provider issues.