Find answers from the community

Updated 2 months ago

Chunk size

Hello, is there a way to define the chunk size for the VectorIndex during its creation? In the past, it could be set via the ServiceContext but now VectorStoreIndex.from_documents(documents, embed_model=my_embd_model) ignores any chunk_size argument.
L
P
2 comments
Plain Text
from llama_index.node_parser import SentenceSplitter

VectorStoreIndex(..., transformations=[SentenceSplitter(chunk_size=512)])
Add a reply
Sign up and join the conversation on Discord