Currently, the Gemini File Search API abstracts the embedding process, providing a seamless but “black-box” indexing experience. With the release of gemini-embedding-2-preview, developers now have access to 8k context windows and native multimodal support.
I would like to know if Google plans to expose a model_name or embedding_config parameter within the file_search_stores.create()method. This would allow us to to enable smart semantic search across images and text natively within the File Search tool on Gemini Calls.