Welcome to Vespa! Following is a list of resources for first time users - Vespa can do many things, and be deployed in many ways. Let us know if something is missing.
|Minimal Quick Start||Install Vespa, insert a few documents and run a query, using Docker or Vagrant. This is the "Hello World" of Vespa, where you can use a laptop or the Vespa Cloud to play with it: Getting Started with Cloud, which includes free trial.|
|Tutorials and Use Cases||
Moving from the minimal quick start to more advanced use casesSearch
Vespa is great for search use cases, but also supports advanced real time online serving of recommender models - see how in the News Search and Recommendation tutorial set.Model deployments
the Models hot swap tutorial describes ways to manage a real-world Vespa application, which is updated on a regular basis while users are performing searches (hot swap). It also extends the application to support multiple recommendation models while minimizing data duplication. Lastly, it demonstrates an efficient garbage collection mechanism, removing obsolete data from the application.Question Answering
From 10 blue links to answering questions. Try the Question Answering use case. This use case demonstrates several of the advanced capabilities of Vespa. Ranking with Transformer (e.g. BERT) models, dense retrieval using vectors, multi-phased retrieval and ranking, and custom plugins to build an end-to-end application.E-Commerce Search
The e-commerce shopping use case demonstrates Vespa grouping, true in-place partial updates, custom ranking and more.
Get started with Vespa python API:
Vespa is an open source platform and can be deployed in multiple ways. These guides are good starters for building multi-node clusters:
Also try deploying to Vespa Cloud, a managed service running Vespa.
Vespa supports writing plugins in languages that runs in the JVM. Use this to simplify query and document processing and build serving logic close to the content nodes, avoid network hops and overhead of HTTP(s):