Customizing Settings #198
chenxizhang
started this conversation in
Use cases
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
We provide some major explicit arguments for the current command, such as
api_key
,endpoint
,model
, etc. However, you can also improve your text generation results through more detailed personalized parameters. You can refer to the official OpenAI API documentation at https://platform.openai.com/docs/api-reference/chat/create to learn about specific parameters, with the most common ones beingmax_tokens
,seed
,stop
,temperature
, etc.Note
Although you can also pass parameters like
messages
,stream
,tools
, etc., it is not recommended because there are implementations within the module for these, and passing these parameters might lead to some unexpected results.You can pass personalized parameters through the
config
parameter (or its aliassettings
), for example:gpt "I need you to help me make a travel plan, a three-day tour of Shanghai, no budget limit. There are children and elderly people, please arrange it to be as comfortable as possible" -config @{max_tokens=1000;temperature=1}
Beta Was this translation helpful? Give feedback.
All reactions