Embeddings with Clarifai

LlamaIndex has support for Clarifai embeddings models.

You must have a Clarifai account and a Personal Access Token (PAT) key. Check here to get or create a PAT.

Set CLARIFAI_PAT as an environment variable.

!export CLARIFAI_PAT=YOUR_KEY

If you’re opening this Notebook on colab, you will probably need to install LlamaIndex 🦙.

!pip install llama-index

Models can be referenced either by the full URL or by the model_name, user ID, and app ID combination.

from llama_index.embeddings import ClarifaiEmbedding

embed_model = ClarifaiEmbedding(
    model_url="https://clarifai.com/clarifai/main/models/BAAI-bge-base-en"
)

# Alternatively
embed_model = ClarifaiEmbedding(
    model_name="BAAI-bge-base-en", user_id="clarifai", app_id="main"
)

embeddings = embed_model.get_text_embedding("Hello World!")
print(len(embeddings))
print(embeddings[:5])