Replies: 2 comments
-
|
Yeah someone recommended using a global setting. I think this would be the best first issue. |
Beta Was this translation helpful? Give feedback.
-
|
Maybe the language localization is not a big issue, at first glance/approximation. Assuming that we do have a powerful LLM (e.g. GPT3.5 and successive) capable of speaking target language X (this is a prerequisite), at first approximation you can continue to use English as the reference language.
The only required modification in prepared prompts is to insert a one-line rule like So you can foresee a global setting variable LANGUAGE set to the target localization language (e.g. equal to To be picky, the above solution has pros and cons:
|
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
This is specially useful for developing countries that don´t speak English. The conversation with the API already works in other languages but the reports don´t.
Beta Was this translation helpful? Give feedback.
All reactions