workman / jiuguan2025cc /src /vectors /makersuite-vectors.js
stail00016's picture
Upload 888 files
4342d5f verified
raw
history blame contribute delete
2.07 kB
import fetch from 'node-fetch';
import { SECRET_KEYS, readSecret } from '../endpoints/secrets.js';
const API_MAKERSUITE = 'https://generativelanguage.googleapis.com';
/**
* Gets the vector for the given text from gecko model
* @param {string[]} texts - The array of texts to get the vector for
* @param {import('../users.js').UserDirectoryList} directories - The directories object for the user
* @returns {Promise<number[][]>} - The array of vectors for the texts
*/
export async function getMakerSuiteBatchVector(texts, directories) {
const promises = texts.map(text => getMakerSuiteVector(text, directories));
return await Promise.all(promises);
}
/**
* Gets the vector for the given text from Gemini API text-embedding-004 model
* @param {string} text - The text to get the vector for
* @param {import('../users.js').UserDirectoryList} directories - The directories object for the user
* @returns {Promise<number[]>} - The vector for the text
*/
export async function getMakerSuiteVector(text, directories) {
const key = readSecret(directories, SECRET_KEYS.MAKERSUITE);
if (!key) {
console.warn('No Google AI Studio key found');
throw new Error('No Google AI Studio key found');
}
const apiUrl = new URL(API_MAKERSUITE);
const model = 'text-embedding-004';
const url = `${apiUrl.origin}/v1beta/models/${model}:embedContent?key=${key}`;
const body = {
content: {
parts: [
{ text: text },
],
},
};
const response = await fetch(url, {
body: JSON.stringify(body),
method: 'POST',
headers: {
'Content-Type': 'application/json',
},
});
if (!response.ok) {
const text = await response.text();
console.warn('Google AI Studio request failed', response.statusText, text);
throw new Error('Google AI Studio request failed');
}
/** @type {any} */
const data = await response.json();
// noinspection JSValidateTypes
return data['embedding']['values'];
}