when creating nodes, how text is

LlamaIndex has default chunking set to 1024 token.
We have default SentenceSplitter that you can go through to understand more.
https://docs.llamaindex.ai/en/stable/module_guides/loading/node_parsers/modules/?h=sentencesplitter#sentencesplitter
https://docs.llamaindex.ai/en/stable/api_reference/node_parsers/sentence_splitter/?h=sentencesplitter#llama_index.core.node_parser.SentenceSplitter