Member Junction
    Preparing search index...

    Interface SemanticTextSegmentationOptions

    Options specific to SemanticTextSegmenter.

    interface SemanticTextSegmentationOptions {
        BlockPreviewChars?: number;
        MaxBlocks?: number;
        MaxSegmentTokens?: number;
        MinSegmentTokens?: number;
        MinTokensForLLM?: number;
        ModelID?: string;
        OverlapTokens?: number;
        PromptName?: string;
    }

    Hierarchy (View Summary)

    Index

    Properties

    BlockPreviewChars? MaxBlocks? MaxSegmentTokens? MinSegmentTokens? MinTokensForLLM? ModelID? OverlapTokens? PromptName?

    Properties

    BlockPreviewChars?: number

    Maximum characters of each block shown to the model. Boundary detection only needs the opening of a block, so truncating keeps the prompt cheap on long documents. Default: 240.

    MaxBlocks?: number

    Maximum blocks sent in one pass. Default: 300.

    MaxSegmentTokens?: number

    Hard ceiling on tokens per text segment. The base class splits any oversized segment via TextChunker so no segmenter can exceed it. Default: 512.

    MinSegmentTokens?: number

    Segments whose text estimates below this many tokens are merged forward into the next segment, preventing a spray of near-empty vectors. Default: 0 (off).

    MinTokensForLLM?: number

    Skip the LLM call entirely when the document estimates below this many tokens — short documents rarely contain multiple topics and the call would not repay its cost. Default: 750.

    ModelID?: string

    Optional model override (an MJ: AI Models ID) for the boundary pass.

    OverlapTokens?: number

    Overlap tokens applied when an oversized segment must be split. Default: 10% of max.

    PromptName?: string

    Name of the MJ: AI Prompts record to run. Default: SEMANTIC_SEGMENTATION_PROMPT_NAME.

    Settings

    Member Visibility

    On This Page

    Properties