The Get Gemini Embeddings process, Pinecone Upsert Embedding process, and Fill Empty Pinecone Index Using Gemini AI process consume the user's input text, modify it to a vector representation, and save it to the Pinecone index.
To configure and publish the processes, perform the following steps:
1Open the following processes in the order specified below:
aGet Gemini Embeddings
bPinecone Upsert Embedding
cFill Empty Pinecone Index Using Gemini AI
dGet Embeddings
eQuery Pinecone
fGenerate Embeddings Model
gQuery LLM With Context (Gemini)
hQuery LLM with Context using Embeddings Model
2For each process, on the Start tab of the Start step, select Cloud Server from the Run On list.
3Optionally, for the Query LLM with Context using Embeddings Model process, in the Set LLM Models step, in the Assignments tab, you can update the values in the Set_Context_Model and Set_Embedding_Model fields.