Rerank intentionally messy candidate documents against a query, then compare the full reranker to the lite version to show the latency-precision tradeoff.
• A valid VOYAGE_API_KEY is set in the environment.
See the exact VAI command, the matching Voyage AI layer, and the MongoDB query shape behind the demo.
vai rerank 'how do I connect to MongoDB Atlas?' --documents 'Use the connection string from your Atlas dashboard' 'Python is a popular language' 'Atlas supports vectorSearch aggregation' 'Copy your URI and pass it to MongoClient' 'The weather in San Francisco is mild'
This tape strips away the vector-search stage so the reranker can be understood on its own. That makes the relevance behavior easier to spot than when reranking is buried inside a bigger retrieval stack.
Share or copy this demo
Keep it lightweight. The prepared text stays behind the buttons.
Share
Copy
LinkedIn opens the share dialog and copies the prepared text so you can paste it in quickly.
The full walkthrough is included here so anyone can replay the demo exactly as published.
More shareable workflows from the same VAI demo library.
Walk through the classic retrieval stack: embed the query, run Atlas vector search, rerank the candidates, then compare the result to a vector-only pass.
VAI command
vai query 'how does vector search work?' --db vai_demo --collection knowledge --model voyage-4-lite
Show Under the Hood
Prerequisites
A valid VOYAGE_API_KEY is set in the environment.
Survey the Voyage model lineup, explain benchmark context, and then measure embedding latency on your own hardware before choosing a production model.
VAI command
vai benchmark embed --input 'What is the best way to index a large document corpus?'
Show Under the Hood
Prerequisites
A valid VOYAGE_API_KEY is set in the environment.
