Installation

1. Install Wagtail Vector Index. You need to specify optional dependencies for vector backend and AI backends (for embeddings and chat models) and it will depend on your project's circumstances.
   • Select an AI provider backend, we recommend starting with LiteLLM.
   • For the vector storage provider, you can choose between:
     • pgvector (Postgres database extension): python -m pip install wagtail-vector-index[pgvector]
     • Qdrant: python -m pip install wagtail-vector-index[qdrant]
     • Weaviate: python -m pip install wagtail-vector-index[weaviate]
     • NumPy: python -m pip install wagtail-vector-index[numpy] (not recommended for big databases or production applications due to scale issues)
     • Read more about storage providers on the specific documentation page: Storage Providers.
   • In your final installation call, you should comma-separate the optional dependencies you want to install, e.g. python -m pip install wagtail-vector-index[litellm,pgvector].
2. Add wagtail_vector_index to your INSTALLED_APPS in your Django project settings file.

   INSTALLED_APPS = [
       # ...
       "wagtail_vector_index",
       # ...
   ]
   

   • If you are using the pgvector backend, you also need to add the specific pgvector app:

     INSTALLED_APPS = [
         # ...
         "wagtail_vector_index",
         "wagtail_vector_index.storage.pgvector",
         # ...
     ]
     

3. Add an AI backend configuration to your Django project settings file. Wagtail Vector Index ships with a backend for the LiteLLM package which you configure for OpenAI in your project settings:

   WAGTAIL_VECTOR_INDEX = {
       "CHAT_BACKENDS": {
           "default": {
               "CLASS": "wagtail_vector_index.ai_utils.backends.litellm.LiteLLMChatBackend",
               "CONFIG": {
                   "MODEL_ID": "gpt-3.5-turbo",
               },
           },
       },
       "EMBEDDING_BACKENDS": {
           "default": {
               "CLASS": "wagtail_vector_index.ai_utils.backends.litellm.LiteLLMEmbeddingBackend",
               "CONFIG": {
                   "MODEL_ID": "text-embedding-ada-002",
               },
           }
       },
   }
   

4. You can supply your OpenAI key in the OPENAI_API_KEY environment variable which is read directly by the openai library (installed by the llm package).
5. Run database migrations.

   ./manage.py migrate