Function: createTextEmbedder()

createTextEmbedder(config, runtime?): Promise<TextEmbedder>

Defined in: extensions/nlp/tasks/textEmbedding.ts:87

Creates a text embedder for executing local Text Embedding models (e.g. sentence-transformers like all-MiniLM-L6-v2).

It loads the tokenizer and model, validates the model input and output requirements, pre-allocates the static execution tensors, and registers clean disposal hooks to clear all native memory. The input text is tokenized and fed at its exact token length (no padding), truncated only when it exceeds the model's maximum sequence length; the attention mask is all ones. Pooling and normalization are baked into the exported .pte; this runner runs the forward pass and returns the raw embedding vector.

Parameters

config

TextEmbedderModel

Text embedder task configuration containing the model and tokenizer paths. See TextEmbedderModel.

runtime?

any

Optional worklet runtime thread on which to run the model execution.

Returns

Promise<TextEmbedder>

A promise resolving to the instantiated TextEmbedder runner.

Throws

With code LOAD_FAILED if the model or tokenizer fails to load, or SCHEMA_MISMATCH if the model schema does not match the text embedding specification.