Replies: 1 comment
-
I would generally advise looking into multilingual models, like the larger Llama models or Command R. If models are multilingual, then German is often one of those languages. Also, I would highly advise simply trying them out and experiment to see whether they fit your particular use case, data, and language. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi, I want to use a local LLM as part of a representation model. The Examples all utilize models primarily used for English. But I have text in German. Can I use models like "zephyr-7b-alpha.Q4_K_M.gguf" with topic models on German texts? Or do I need to use a LLM specifically trained for German? And if so, which model should I use?
I already did a search and found models such as "TheBloke/leo-hessianai-7B-chat-GGUF" or "Nekochu/Llama-3.1-8B-German-ORPO". But perhaps there are other LLMs more appropriate?
Beta Was this translation helpful? Give feedback.
All reactions