Welcome to Marqo Docs!
Marqo (pronounced "mar-co") is a unified embedding generator and search engine.
This documentation will provide you with the information needed to get started with Marqo and to get the most out of its features!
New to Marqo?
Get started with this Quickstart Tutorial and learn how to build your first vector search project with Marqo.
Questions
Ask your questions in our Community Slack where the Marqo Team and our amazing community can help.
Step 1: Select Your Use Case
Marqo is a fully-fledged vector search solution which means it can be used in a variety of different scenarios. Such scenarios include but are not limited to:
Text Search
Search and retrieve relevant textual data using Marqo's powerful text search capabilities.
Image Search
Find and identify images effortlessly using Marqo's intelligent image search engine.
RAG
Combine retrieval-augmented generation with Marqo to generate precise and context-aware responses.
Step 2: Choose Your Deployment
Click on the items below to find the platform most suited to you.
Marqo Cloud
- From evaluation to production
- Zero-downtime scale up and down with the click of a button
- Hundreds of open source base models or the ability to bring your own
- Fine-tune your models using our embedding training platform, Marqtune
- 24/7 support
- Datadog integration
- Zero-downtime updates
- (Optional) high availability
Open Source (Docker)
- For local evaluation & development
- Hundreds of open source base models or the ability to bring your own
- Great for POCs
- Try out the latest features
- GPU support
- Provide personal feedback
Step 3: Start Building!
Once you've decided on your deployment and use case, navigate to the relevant sections to get started. Below are some resources we'd recommend checking out.