Skip to content

Using GPT-4 #156

Answered by mspronesti
wenger9 asked this question in Q&A
May 22, 2023 · 1 comments · 2 replies
Discussion options

You must be logged in to vote

just set the model parameter in the OpenAI class when instiating it, e.g.

from pandasai import PandasAI
from pandasai.llm import OpenAI

llm = OpenAI(api_key='YOUR_KEY_HERE', model='gpt-4')
pandas_ai = PandasAI(llm, conversational=False)
pandas_ai(df, prompt='YOUR_PROMPT_HERE')

if you use AzureOpenAI just create a deployment of GPT-4 and set deployment_name to the deployment name.

Hope this helps! If it does, please accept my answer 😄

Replies: 1 comment 2 replies

Comment options

You must be logged in to vote
2 replies
@wenger9
Comment options

@mspronesti
Comment options

Answer selected by wenger9
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants