llama-index-embeddings-ibm

What it is and what it does

This package bridges LlamaIndex's embedding interface with IBM watsonx.ai's foundation models, letting you use IBM's embedding models (such as slate-125m-english-rtrvr-v2) within LlamaIndex workflows. It wraps the ibm-watsonx-ai SDK to handle authentication, model selection, and embedding computation.

You initialize WatsonxEmbeddings with your IBM Cloud credentials, project context, and model choice, then call get_query_embedding() for single strings or get_text_embedding_batch() for multiple texts. The package handles token truncation and API communication with watsonx.ai, making it straightforward to swap IBM models into existing LlamaIndex pipelines that expect an embeddings provider.

Use it for:

• Embed documents and queries using IBM's models within a LlamaIndex retrieval-augmented generation (RAG) pipeline.
• Build semantic search applications that leverage IBM watsonx.ai's multilingual or domain-specific embedding models.
• Integrate IBM's foundation models into LlamaIndex-based applications already using other LLM providers.
• Batch embed large text collections for indexing with LlamaIndex using IBM's infrastructure.
• Evaluate IBM embedding models as an alternative provider within existing LlamaIndex applications.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Integrates IBM watsonx.ai embedding models with LlamaIndex, allowing you to embed text strings using IBM's foundation models through the LlamaIndex API.

Yes, if you are already using LlamaIndex and have access to IBM watsonx.ai. The package has low install friction, active maintenance, no security vulnerabilities, and a permissive license. It is most valuable for teams committed to IBM's cloud platform or evaluating IBM's embedding models; it adds little value if you lack watsonx.ai access or prefer other embedding providers.

Install

llama-index-embeddings-ibm on PyPI

pip install llama-index-embeddings-ibm

uv add llama-index-embeddings-ibm

poetry add llama-index-embeddings-ibm

Installing llama-index-embeddings-ibm

Before you install

Low friction install with three runtime dependencies. Actively maintained as of February 2026 with no known vulnerabilities.

License in practice

MIT license permits commercial and private use with minimal restrictions.

Quickstart

pip install llama-index-embeddings-ibm

from llama_index.embeddings.ibm import WatsonxEmbeddings

watsonx_embedding = WatsonxEmbeddings(
    model_id="ibm/slate-125m-english-rtrvr-v2",
    url="https://us-south.ml.cloud.ibm.com",
    project_id="YOUR_PROJECT_ID"
)

query_result = watsonx_embedding.get_query_embedding("Example query.")

Requires IBM Cloud API key (WATSONX_APIKEY environment variable), project_id or space_id, and network access to IBM watsonx.ai service.

Verify before relying

• Performance characteristics and latency of embedding operations compared to other providers.
• Supported embedding model list and their respective dimensions/capabilities.
• Rate limiting and quota behavior under production load.

Package facts

Evidence: llama_index_embeddings_ibm-0.6.0.post1-py3-none-any.whl

Tags

More Artificial Intelligence packages