MistralAI Embeddings

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

%pip install llama-index-embeddings-mistralai

!pip install llama-index

# imports
from llama_index.embeddings.mistralai import MistralAIEmbedding

# get API key and create embeddings
api_key = "YOUR API KEY"
model_name = "mistral-embed"
embed_model = MistralAIEmbedding(model_name=model_name, api_key=api_key)

embeddings = embed_model.get_text_embedding("La Plateforme - The Platform")

print(f"Dimension of embeddings: {len(embeddings)}")

Dimension of embeddings: 1024

embeddings[:5]

[-0.0299224853515625,
 -0.0028362274169921875,
 0.0282745361328125,
 -0.034759521484375,
 -0.0017366409301757812]