6
7```
8<a href="https://www.val.town/settings/environment-variables?name=OpenAI&value=sk-123...">
9 Add OpenAI key to Val Town
10</a>
11```
1import { email } from "https://esm.town/v/std/email?v=11";
2import { OpenAI } from "npm:openai";
3import { createHeaders, PORTKEY_GATEWAY_URL } from "npm:portkey-ai";
4
9).then(r => r.json());
10
11const openai = new OpenAI({
12 baseURL: PORTKEY_GATEWAY_URL,
13 defaultHeaders: createHeaders({
14 apiKey: Deno.env.get("PORTKEY_API_KEY"),
15 virtualKey: Deno.env.get("PORTKEY_OPENAI_VIRTUAL_KEY"),
16 }),
17});
18let chatCompletion = await openai.chat.completions.create({
19 messages: [{
20 role: "user",
2import { z } from "npm:zod";
3
4// Function to handle image and text input using OpenAI's GPT-4-turbo
5async function handleImageChat() {
6 const initialMessages = [
13 const response = await modelProvider.gen({
14 model: "gpt-4-turbo",
15 provider: "openai",
16 messages: [
17 ...initialMessages,
1import { OpenAI } from "https://esm.town/v/std/openai";
2
3const openai = new OpenAI();
4const completion = await openai.chat.completions.create({
5 "messages": [
6 {
1import { email } from "https://esm.town/v/std/email?v=11";
2import { OpenAI } from "npm:openai";
3
4let location = "toronto on";
8).then(r => r.json());
9
10const openai = new OpenAI();
11let chatCompletion = await openai.chat.completions.create({
12 messages: [{
13 role: "user",
6import cosSimilarity from "npm:cos-similarity";
7import _ from "npm:lodash";
8import OpenAI from "npm:openai";
9
10const dimensions = 1536;
27 await Promise.all(allBatchDataIndexesPromises);
28
29 const openai = new OpenAI();
30 const queryEmbedding = (await openai.embeddings.create({
31 model: "text-embedding-3-small",
32 input: query,
3import { sqlToJSON } from "https://esm.town/v/nbbaier/sqliteExportHelpers?v=22";
4import { db as allValsDb } from "https://esm.town/v/sqlite/db?v=9";
5import OpenAI from "npm:openai";
6
7export default async function semanticSearchPublicVals(query) {
11 });
12
13 const openai = new OpenAI();
14
15 const embedding = await openai.embeddings.create({
16 model: "text-embedding-3-small",
17 input: query,
3Uses [Neon](https://neon.tech/) to search embeddings of all vals, using the [pg_vector](https://neon.tech/docs/extensions/pgvector) extension.
4
5- Call OpenAI to generate an embedding for the search query.
6- Query the `vals_embeddings` table in Neon using the cosine similarity operator.
7 - The `vals_embeddings` table gets refreshed every 10 minutes by [janpaul123/indexValsNeon](https://www.val.town/v/janpaul123/indexValsNeon).
1*Part of [Val Town Semantic Search](https://www.val.town/v/janpaul123/valtownsemanticsearch).*
2
3Generates OpenAI embeddings for all public vals, and stores them in [Neon](https://neon.tech/), using the [pg_vector](https://neon.tech/docs/extensions/pgvector) extension.
4
5- Create the `vals_embeddings` table in Neon if it doesn't already exist.
6- Get all val names from the [database of public vals](https://www.val.town/v/sqlite/db), made by [Achille Lacoin](https://www.val.town/u/pomdtr).
7- Get all val names from the `vals_embeddings` table and compute the difference (which ones are missing).
8- Iterate through all missing vals, get their code, get embeddings from OpenAI, and store the result in Neon.
9- Can now be searched using [janpaul123/semanticSearchNeon](https://www.val.town/v/janpaul123/semanticSearchNeon).
10
6import cosSimilarity from "npm:cos-similarity";
7import _ from "npm:lodash";
8import OpenAI from "npm:openai";
9
10const openai = new OpenAI();
11const queryEmbedding = (await openai.embeddings.create({
12 model: "text-embedding-3-small",
13 input: "check dynamicland website for changes and email me",
16console.log(queryEmbedding.slice(0, 4));
17
18const embedding = await openai.embeddings.create({
19 model: "text-embedding-3-small",
20 input: "check dynamicland website for changes and email me",
41}`;
42
43const queryEmbeddingVal = (await openai.embeddings.create({
44 model: "text-embedding-3-small",
45 input: valCode,
