Crate ck_chunk

Crate ck_chunk 

Source

Structs§

Chunk
ChunkConfig
Configuration for chunking behavior
ChunkMetadata
StrideInfo
Information about chunk striding for large chunks that exceed token limits
TokenEstimator
Import token estimation from ck-embed Simple token estimation for code and text This is a rough approximation since we don’t have access to the actual model tokenizer

Enums§

ChunkType
ParseableLanguage

Functions§

chunk_text
chunk_text_with_config
chunk_text_with_model
New function that accepts model name for model-specific chunking
get_model_chunk_config
Get model-specific chunk configuration (target_tokens, overlap_tokens) Balanced for precision vs context - larger models can handle bigger chunks but not too big