-
Navigate to the directory where the repository was downloaded
cd snowflakegpt
-
Install the required dependencies
pip install -r requirements.txt
-
Configure OpenAI Key
- If Using OpenAI key, simply
export OPENAI_API_KEY=*****
- If want to use config file, rename
config_template.ini
->config.ini
file inside thesnowdflakegpt
dir & update either Azure or OpenAI config
By completing these steps, you have properly configured the API Keys for your project.
- If Using OpenAI key, simply
- Run
snowdflakegpt
Python module in your terminal with default GPT model
python -m snowgpt "What is the population of India?"
- Run the agent by passing specific model.
python -m snowgpt "How many women between the ages 39 and 45 in India" gpt-4_8k_ascent
Current supported list of deployment names are
- gpt-35-turbo
- gpt-4_8k_ascent
- gpt-4_32k_ascent
Default is gpt-35-turbo
Maybe a sequential chain to have an intermediate function to validate the query and make corrections
Creating a Custom Agent to control the query generation. This may need a custom Agent & custom Tools .