mirror of
https://github.com/abetlen/llama-cpp-python.git
synced 2023-09-07 17:34:22 +03:00
`llama.create_chat_completion` definitely has a `top_k` argument, but its missing from `CreateChatCompletionRequest`. decision: add it