Problem with Textgeneration

#3
by doomgrave - opened

Hi. im using Textgeneration + llama.cpp + OpenAI api.
Whatever setting i use I get this error:
$This model maximum context length is 2048 tokens. However, your messages resulted in over 1021 tokens and max_tokens is 2048.

But it seems the model should be 4096. May I must set some special parameter?
n_ctx it's already 4096

Sign up or log in to comment