Sign in

Late Chunker

Implements the late chunking strategy. Encodes the entire text, then splits recursively, deriving chunk embeddings from the full document embedding.

post/v1/chunk/late

Response

Successful Response: A list of LateChunk objects.

textstring

The actual text content of the chunk.

start_indexinteger

The starting character index of the chunk within the original input text.

end_indexinteger

The ending character index (exclusive) of the chunk within the original input text.

token_countinteger

The number of tokens in this specific chunk, according to the tokenizer used.

embeddingnumber[] nullable

Optional embedding vector (list of floats) for the entire chunk, derived from the full document embedding.

Changes

Changed in 13 of the 20 revisions of this API.131925

  • 8a6665f478db1See the full diff
    • ●

      removed the optional property items/sentences from the response with the 200 status

      response-optional-property-removed

  • bc64e17759521See the full diff
    • ○

      endpoint added

      endpoint-added

    This revision also has 1 change that name no endpoint, such as unreferenced schemas being removed. See the revision's changelog

  • 5c8e448048061See the full diff
    • ▲

      api removed without deprecation

      api-removed-without-deprecation

    This revision also has 19 changes that name no endpoint, such as unreferenced schemas being removed. See the revision's changelog

  • 1762a3ecb81b1See the full diff
    • ○

      the embedding_model request property default value changed from sentence-transformers/all-minilm-l6-v2 to all-MiniLM-L6-v2

      request-property-default-value-changed

  • 7e3cb743a9ed1See the full diff
    • ●

      removed the request property text

      request-property-removed

  • b5fe83cf70c611See the full diff
    • ▲

      removed the media type application/json from the request body

      request-body-media-type-removed

    • ○

      added the media type multipart/form-data to the request body

      request-body-media-type-added

  • e779d3fd51d512See the full diff
    • ●

      removed the request property rules

      request-property-removed

    • ○

      added the new optional request property lang

      new-optional-request-property

    • ○

      added the new optional request property recipe

      new-optional-request-property

  • 4c7520e794a81See the full diff
    • ○

      the min_characters_per_chunk request property default value changed from 1 to 24

      request-property-default-value-changed

  • d612ec62c8f662See the full diff
    • ●

      removed the request property approximate

      request-property-removed

    • ●

      removed the request property delim

      request-property-removed

    • ●

      removed the request property include_delim

      request-property-removed

    • ●

      removed the request property min_characters_per_sentence

      request-property-removed

    • ●

      removed the request property min_sentences_per_chunk

      request-property-removed

    • ●

      removed the request property mode

      request-property-removed

    • ○

      added the new optional request property min_characters_per_chunk

      new-optional-request-property

    • ○

      added the new optional request property rules

      new-optional-request-property

  • 2d174e4e5a5a1See the full diff
    • ○

      the request property chunk_size became optional

      request-property-became-optional