learning-it by example

Back to the video

pgvector vs Chroma vs Qdrant: Same Data, Real Results

Same notes, same question, three vector databases on your laptop: pgvector, Chroma and Qdrant, with real output and three classic mistakes.

Video #03 ยท RAG and vector search

Steps in the video

  1. 0:00 Which vector database?
  2. 0:19 Three ways to store vectors
  3. 0:55 Setup
  4. 1:11 Two containers
  5. 1:40 The same notes in three stores
  6. 2:56 Which one should you pick
  7. 3:27 Three classic mistakes
  8. 4:37 What this does not show
  9. 5:11 Four mistakes to avoid
  10. 5:31 My take and next steps
  11. 5:58 Recap

Files

  • common.py (26 lines)
  • compare.py (65 lines)
  • compose.yaml (16 lines)
  • gotchas.py (58 lines)
  • notes.txt (22 lines)

Download the code (zip)

Tested with

Python 3.12.15 | ollama==0.6.3, psycopg[binary]==3.3.6, pgvector==0.5.1, chromadb==1.5.9, qdrant-client==1.20.0

What this does not show

One question over seven made-up notes on one laptop; no test of speed or scale. The query-plan result is from 10,000 random vectors with an HNSW index. Behaviour is for the pinned versions. Which-to-pick advice is from blog posts, not a benchmark.

Sources

Learning IT by Example.