Snippet: Batch create embedding vectors

Batch create embedding vectors

OpenAI — openai

Efficiently creates embedding vectors by batching them in requests using the embeddings/create API

const lib = require('lib')({token: process.env.STDLIB_SECRET_TOKEN});

// Array of input
let inputs = context.params.inputs || [];
let embeddings = [];

// batch inputs so we don’t exceed token limits
// tokens aren’t exactly words, so we’ll limit tokenCount to 4096
// in case of weird characters, this should handle most input variatons
while (inputs.length) {
  let tokenCount = 0;
  let batchedInputs = [];
  while (inputs.length && tokenCount < 4096) {
    let input = inputs.shift();
    batchedInputs.push(input);
    tokenCount += input.split(' ').length;
  }
  let embeddingResult = await lib.openai.playground['@0.0.4'].embeddings.create({
    model: `text-embedding-ada-002`,
    input: batchedInputs
  });
  embeddings = embeddings.concat(
    embeddingResult.data.map(entry => entry.embedding)
  );
}

return embeddings;