Lesson 82 · 2 min
Prompt in Vietnamese or English?
Write in the language you think in, and ask for the answer in the language you need.
Big chatbots handle Vietnamese well for everyday tasks: emails, explanations, summaries, ideas. You do not need English to get a useful answer.
English can still help in a few cases. Models saw far more English text while training, so very technical or niche topics sometimes come out richer in English. Vietnamese text also tends to use more tokens, the small pieces models read, so a very long document may fill a chat's working memory a little sooner.
You can mix. Write your prompt in Vietnamese, keep technical terms in English where your field uses them, and say "Answer in Vietnamese" or "Answer in English" at the end. The model follows that instruction well.
Check the Vietnamese it writes. Diacritics, word choice and how polite it sounds can be slightly off, and it may use northern or southern words that do not fit your readers.
Do it now
Try the same request both ways
- 1. Pick a question from your work or studies.
- 2. Ask it once in Vietnamese and once in English, both times asking for the answer in Vietnamese.
- 3. Compare depth, accuracy and how natural the Vietnamese sounds.
[YOUR QUESTION] Answer in Vietnamese, in a friendly everyday tone, for readers in [CITY OR REGION]. Keep these terms in English: [TERMS].
Quick check
Keep going
Next →Ask for options, then chooseWords from this lesson
- Token
- A chunk of text, often part of a word, that AI reads and writes in. Limits and prices are counted in tokens.
- Training data
- The examples, such as text, images or recordings, that a model learns its patterns from.
- LLM
- An AI trained on huge amounts of text to predict the next word, which lets it write, summarise, translate and chat.