The community member is asking if it is possible to use multiple embedding models for information retrieval, take the top k results from each model, and then use those results for predictions with the LlamaIndex. The comments suggest that the community member may be looking to use an ensemble retrieval approach, and provide a link to the LlamaIndex documentation on this topic, which may be helpful.
hi. Can we use multiple embedding models for information retrieval and take top k from each model and then send those results for predictions using llama Index ?
