LlamaIndex

IBM integration with LlamaIndex is available under IBM and LlamaIndex LLM integration documentation, while for embedding integration you may visit IBM and LlamaIndex Embedding integration documentation.

Note

The llama-index-llms-ibm==0.2.x package relies on version 0.11.x of the llama-index-core package and ibm_watsonx_ai in version >=1.1.x.

Installation and Setup

Install the integration package with

pip install -qU llama-index-llms-ibm

Get an IBM watsonx.ai api key and set it as an environment variable (WATSONX_APIKEY)

import os

os.environ["WATSONX_APIKEY"] = watsonx_api_key

LLMs

See detailed example how to load a model in the following link

from llama_index.llms.ibm import WatsonxLLM

Embedding Models

Install the integration package with

pip install -qU llama-index-embeddings-ibm

Note

The llama-index-embeddings-ibm==0.1.x package relies on version 0.11.x of the llama-index-core package and ibm_watsonx_ai in version >=1.0.x.

Usage example:

from llama_index.embeddings.ibm import WatsonxEmbeddings

More details can be found here

API reference

For detailed documentation of all IBM watsonx.ai features and configurations head to the API reference