Skip to content
BENIBIX Learn

Chroma (Vector Database)

AI🔎 Retrieval & SearchVector Databases

Also known as ChromaDB, Chroma DB

A free vector database that anyone can use. You can start in minutes.

Think of it like

It is like a notebook on your own desk. You can start right away.

Example

Emma tests a RAG idea on her laptop. She runs pip install chromadb, adds 300 notes and asks questions within ten minutes.

Why it matters

It lets you try retrieval in minutes. You find out if RAG helps before you build anything big.

Definition

Chroma is a free, open-source vector database that is easy to start with. You install it with one command and use it from Python or JavaScript. You hand it texts, and it can make the embeddings for you. Then you ask it a question, and it returns the closest texts. It can run inside your app on a laptop, or as a server. Many RAG tutorials begin with it.

Example code

python
import chromadb
notes = chromadb.Client().create_collection("notes")
notes.add(documents=["The gym opens at 6am."], ids=["n1"])
notes.query(query_texts=["When does the gym open?"], n_results=1)
Add one note, then ask a question in other words and get that note back.
You run Chroma yourself. It can even run on a laptop.

Common mistake

Keeping the laptop setup for real users. Plan how it will run, back up and grow before you launch.

Key takeaways

  1. 1
    Quick start. One install, a few lines.
  2. 2
    Makes embeddings. Hand it text, not vectors.
  3. 3
    Laptop or server. Same code, two ways to run.