Great question — answering as the UphillTech team:
This is an issue when running in Mac with Intel chips. You can solve this by taking these steps: Change the onnxruntime version: !pip install onnxruntime==1.16.3 and install sentence_transformers: !pip install sentence_transformers Now, you create the embedding separately: from chromadb.utils import embedding_functions sentence_transformer_ef = embedding_functions.SentenceTransformerEmbeddingFunction(model_name="all-MiniLM-L6-v2") and then loop through your documents and upsert separately: documents = ["This is document1", "This is document2"] metadatas = [{"source": "notion"}, {"source": "google-docs"}] ids = ["doc1", "doc2"] for doc, meta, id in zip(documents, metadatas, ids):...
Full answer by GeekyGeek: https://stackoverflow.stackexchange.com/a/78911714 (CC BY-SA 4.0).
Need it done for your business? UphillTech.
