Find answers from the community

Updated last year

Local

At a glance

The community member is looking for recommendations to use LlamaIndex without any OpenAI API key or embeddings, index, or vector data from OpenAI. In the comments, another community member suggests using open-source language models and embedding models instead of OpenAI, and persisting all data locally. They provide links to the LlamaIndex documentation for using local LLMs and embedding models. However, the last link they provided for setting the service context globally does not work.

Useful resources

aasdw2.

any recommendations to do Discover LlamaIndex: Bottoms-Up Development With LLMs without openapi key usage at all? no embeddings/index/vector nothing from openai to be used in llama setup/dev

2 comments

WWhiteFang_Jr

You can use open source llm and embed model in place of openAI.

Persist all the data locally.
For local llm: https://docs.llamaindex.ai/en/stable/understanding/using_llms/using_llms.html#using-a-local-llm

For local embed model: https://docs.llamaindex.ai/en/stable/module_guides/models/embeddings.html#local-embedding-models

Declare the service context global: https://docs.llamaindex.ai/en/stable/module_guides/supporting_modules/service_context.html#setting-global-configuration

RRamin

The last link (service context global) doesn't work @WhiteFang_Jr

Add a reply