LLM Text Completion

• clone this repo

• rename example.env to .env

• set environment variables in .env file

  • OPENAI_API_KEY from https://platform.openai.com/account/api-keys
  • Azure OpenAI Service Settings from Azure OpenAI https://portal.azure.com
    • AZURE_OPENAI_KEY
    • AZURE_OPENAI_ENDPOINT
    • AZURE_OPENAI_DEPLOYMENT_NAME
  • GOOGLE_PALM_AI_API_KEY from https://makersuite.google.com
  • LLaMA-2 Together API
    • TOGETHER_API_KEY from https://api.together.xyz/playground,
    • Start LLaMA-2 Chat (70B)
  • Vertex AI
    • GOOGLE_PROJECT_ID from Google Cloud console, refer to this https://cloud.google.com/vertex-ai/docs/start/cloud-environment
    • In the Google Cloud console, on the project selector page, select or create a Google Cloud project.
      • Go to project selector
    • Make sure billing is enabled for your Google Cloud project.
    • Enable the Vertex AI API
      • Enable the API.
    • Add GCP Credential file as gcp-cred.json for Vertex AI, IAM -> Service accounts -> An account -> Keys from https://console.cloud.google.com/iam-admin/serviceaccounts.

• for running individual Python programs, use this

  Installation

  pip install -r requirements.txt
  

  OpenAI Python API

  python text-completion.py
  

  Azure OpenAI Service

  python text-completion-azure.py
  

  LLaMA-2 Chat (70B)

  python text-completion-together-api-llama.py
  

  Vertex AI

  python text-completion-vertex-ai.py
  

or use these notebooks

• Gradio App
• OpenAI text-completion
• Azure OpenAI text-completion

Further, simplify using LangChain or Semantic Kernel

• Text Completion via LangChain
• Text Completion via Semantic Kernel

ChatGPT Plugin

• Plugins/examples
• ChatGPT plugins quickstart
• ChatGPT Retrieval Plugin
• Awesome ChatGPT Plugins

More comprehensive demos are available on LLM Scenarios, Use cases on the Gradio app