LlamaIndex Embeddings Integration: Oracle Cloud Infrastructure (OCI) Generative AI

Oracle Cloud Infrastructure (OCI) Generative AI is a fully managed service that provides a set of state-of-the-art, customizable large language models (LLMs) that cover a wide range of use cases, and which are available through a single API. Using the OCI Generative AI service you can access ready-to-use pretrained models, or create and host your own fine-tuned custom models based on your own data on dedicated AI clusters.

Installation

bash

pip install llama-index-embeddings-oci-genai

You will also need to install the OCI sdk

bash

pip install -U oci

Usage

bash

from llama_index.embeddings.oci_genai import OCIGenAIEmbeddings

embedding = OCIGenAIEmbeddings(
    model_name="MY_MODEL",
    service_endpoint="https://inference.generativeai.us-chicago-1.oci.oraclecloud.com",
    compartment_id="MY_OCID",
)