Chunking

Late Chunk

Late chunking endpoint using Modal.

Chunks text using late chunking approach with embeddings computed at chunk boundaries. This endpoint calls the Modal function for GPU-accelerated processing.

post/v1/chunk/late

Headers

authorizationstring nullable

Request body

ocrstring

OCR provider (mistral or gemini)

embedding_modelstring

Embedding model

chunk_sizeinteger

Size of each chunk

recipestring

Chunking recipe to use

langstring

Language code

min_characters_per_chunkinteger

Minimum characters per chunk

Response

Successful Response

{"stackTrail":"paths:/v1/chunk/late:post:responses:200:content:application/json:schema","oasType":"schema","type":"unknown"}

Changes

No recorded changes to this endpoint across all 1 revision of this API.