/shuttle-qdrant-template

Primary LanguageRust

Rust, Qdrant, Shuttle: A Rust-based semantic document search

This example shows how to create a semantic document search with chat capability in Rust, using Shuttle and Qdrant - a vector search engine.

Prerequisites

  1. Install Shuttle CLI tools:
cargo install cargo-shuttle

  1. Get a Qdrant Account at qdrant.tech

  2. Get an OpenAI API key at openai.com

  3. Create a Secrets.toml file (or Secrets.dev.toml for local development) with the following contents:

OPENAI_API_KEY = "<ENTER YOUR KEY HERE>"
QDRANT_TOKEN = "<ENTER YOUR KEY HERE>"
QDRANT_URL = "<ENTER YOUR KEY HERE>"

Running the example

cargo shuttle run

Troubleshooting

  • If you run the service but it only lists the files loaded and doesn't try to embed them, check you have credits/funding in your OpenAI account.
  • Make sure your Qdrant URL ends with port :6334.