# Vector Databases

**URL:** <https://replit.discourse.group/t/vector-databases/1049>\
**Category:** Tips & Tricks\
**Tags:** how-to\
**Created:** [January 20, 2025, 4:15am UTC](https://replit.discourse.group/t/vector-databases/1049 "2025-01-20T04:15:40Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![preston23](https://avatars.discourse-cdn.com/v4/letter/p/c0e974/32.png) [@preston23](https://replit.discourse.group/u/preston23)\
**Post date:** [January 20, 2025, 4:15am UTC](https://replit.discourse.group/t/vector-databases/1049/1 "2025-01-20T04:15:40Z")

</div>

Is a Vector database possible on Replit? Or do I need to hook into Supabase?

---

<div class="post-metadata">

**Author:** ![nobita2041](https://yyz2.discourse-cdn.com/flex010/user_avatar/replit.discourse.group/nobita2041/32/157_2.png) [@nobita2041](https://replit.discourse.group/u/nobita2041)\
**Post date:** [January 20, 2025, 10:18am UTC](https://replit.discourse.group/t/vector-databases/1049/2 "2025-01-20T10:18:19Z")

</div>

I do not know how to use vector databases.  
However, I have found that neondb supports pgvector.  
And it could actually be enabled in the Replit database.  
Below is the result of enabling vecor from the tools shell.

```auto
~/workspace$ psql
psql (16.5, server 17.2)
WARNING: psql major version 16, server major version 17.
         Some psql features might not work.
SSL connection (protocol: TLSv1.3, cipher: TLS_AES_256_GCM_SHA384, compression: off)
Type "help" for help.

neondb=> CREATE EXTENSION vector;
CREATE EXTENSION
Quit (core dumped)
~/workspace$ 

```

Also, the following instructions for using pgvector with neon might help you.

> **[The pgvector extension - Neon Docs](https://neon.tech/docs/extensions/pgvector)**
>
> The pgvector extension enables you to store vector embeddings and perform vector similarity search in Postgres. It is particularly useful for applications involving natural language processing, such a...

---

<div class="post-metadata">

**Author:** ![kody-replit](https://yyz2.discourse-cdn.com/flex010/user_avatar/replit.discourse.group/kody-replit/32/3_2.png) [@kody-replit](https://replit.discourse.group/u/kody-replit)\
**Post date:** [January 20, 2025, 10:53pm UTC](https://replit.discourse.group/t/vector-databases/1049/3 "2025-01-20T22:53:18Z")

</div>

The neondb postgres supports pgvector, Agent might be able to singleshot it if you provide docs/context but it’s not a specific tool for the Agent right now

---

<div class="post-metadata">

**Author:** ![ztalexander](https://avatars.discourse-cdn.com/v4/letter/z/47e85d/32.png) [@ztalexander](https://replit.discourse.group/u/ztalexander)\
**Post date:** [January 22, 2025, 1:22am UTC](https://replit.discourse.group/t/vector-databases/1049/4 "2025-01-22T01:22:39Z")

</div>

I’ve been using Pinecone vector database with Voyage-2 and Voyage-3 embeddings. The biggest issues I’ve found is that Voyage-2 and Voyage-3 use a 1024 vector size. On the other hand, OpenAI likes vector sizes for 1536 and larger. Pinecone is considered the top vector database. And you can learn it for free. Voyage-3 requires a credit card. However, you 200 Million free tokens.  
–Zachary

---

<div class="post-metadata">

**Author:** ![jason392](https://avatars.discourse-cdn.com/v4/letter/j/5f9b8f/32.png) [@jason392](https://replit.discourse.group/u/jason392)\
**Post date:** [February 1, 2025, 11:31am UTC](https://replit.discourse.group/t/vector-databases/1049/5 "2025-02-01T11:31:52Z")

</div>

> [@ztalexander](#):
>
> OpenAI

pg vector defintly lets you do larger vectors than 1000. I think for a small use case postgres is fine.

---

<div class="post-metadata">

**Author:** ![ztalexander](https://avatars.discourse-cdn.com/v4/letter/z/47e85d/32.png) [@ztalexander](https://replit.discourse.group/u/ztalexander)\
**Post date:** [February 1, 2025, 3:30pm UTC](https://replit.discourse.group/t/vector-databases/1049/6 "2025-02-01T15:30:01Z")

</div>

1024 vs. 1536 is more of a cost-saving issue. Obviously, 1024 takes up less space in GPU memory. It doesn’t stop there you should also look at the embeddings. I have been using Voyage-3 Embeddings. However, I’ve switched to Voyage-3-large and added rerank-lite-1.

You always have a choice. I choose to use tools that I would probably use in production. I want to go as deep as possible on my chosen tech stack. Speed matters when money is on the line. And I don’t want to be learning how to use a new tool when costs count. I’d rather implement something new using tools I’m already comfortable using.