What is the precision of a float in vector?

Hello! I’m trying to understand how much storage space is in 100,000 vectors. Does anyone know what precision() is used for each float in a pinecone vector?

Hey @byzan.solutions, each dimension on a single vector consumes 4 bytes of memory and storage per dimension. See: “Dimensionality of vectors” section in Choosing index type and size

Hey @byzan.solutions , more specifically we use 32-bit floats as defined in IEEE 754, which I think is basically the industry standard at this point (Like Java’s float type, or Rust’s f32).