LlamaIndex
Log in

Find answers from the community

Updated 2 years ago

is there a model that can output more

ddx31
·
is there a model that can output more than 512 tokens? im trying to summarize huge docs and it seems to be defaulting to 512 tokens (i want the token output to be around 2000)
b
d
4 comments
bbbornsztein
You can configure the max_tokens parameter: https://langchain.readthedocs.io/en/latest/_modules/langchain/llms/openai.html?highlight=max_tokens#

Plain Text
llm = OpenAI(max_tokens=2000)
ddx31
I've tried this and it doesn't work.
bbbornsztein
See here: https://gpt-index.readthedocs.io/en/latest/how_to/custom_llms.html#example-fine-grained-control-over-all-parameters
bbbornsztein
is that what you tried?
Add a reply
Sign up and join the conversation on Discord