LunaTranslator/docs/en/offlinellm.md

How to Use Large Model Offline Translation?

ChatGPT Compatible Interface

You can also use tools like llama.cpp, ollama, one-api to deploy models, and then fill in the address and model.

You can also use platforms like Kaggle to deploy models to the cloud, in which case you may need to use SECRET_KEY; otherwise, you can ignore the SECRET_KEY parameter.

You can also fill in the API of the registered large model (but it's unnecessary), and compared to the ChatGPT compatible interface under registered online translation, the only difference is that it does not use a proxy.