This function sets up an large language model API for tasks.
Usage
setupAgent(
name = c("openai", "replicate", "generic"),
type = NULL,
model = NULL,
url = NULL,
ai_api_key = Sys.getenv("AI_API_KEY")
)Arguments
- name
A string for the name of the API, one of "openai", "replicate" or "generic". Currently supported APIs are "openai" and "replicate". If the user wishes to use another API that has similar syntax to openai API this is also supported via the the "generic" option. In this case, the user should also provide a url for the API using the
- type
Specify type of model (chat or completion). This parameter only needs to be specified when using 'openai
- model
LLM model you wish to use. For openAI chat model examples are:
'gtp-3-5-turbo'
'gtp-4'
For openAI completion models examples are:
'text-curie-001'
'text-davinci-002'
For replicate models examples are:
llama-2-70b-chat ( as '02e509c789964a7ea8736978a43525956ef40397be9033abf9fd2badfe68c9e3')
llama-2-13b-chat ( as 'f4e2de70d66816a838a89eeeb621910adffb0dd0baba3976c96980970978018d')
For a full list of openAI models see https://platform.openai.com/docs/models/overview/. For a full list of Replicate models, see https://replicate.com/collections/language-models.
- url
the url for the API in case the API "generic" is selected. (Default: NULL)
- ai_api_key
personal API key for accessing LLM
Examples
{
myAgent <- setupAgent(name="openai",type="chat",model="gpt-4",ai_api_key="my_key")
myAgent <- setupAgent(name="replicate",type=NULL,
model="02e509c789964a7ea8736978a43525956ef40397be9033abf9fd2badfe68c9e3",
ai_api_key="my_key")
}
#> Warning: Request for available models failed. Check your API-key.