Create New Benchmark

Set up a conversation and assertions for your AI agent test

Basic Information

Provide a name and database for your benchmark

Benchmark Name

Database

All benchmarks will run in the selected databaseLoading databases...

Collections (Optional)

Loading collections...

Optional: Add this benchmark to one or more collections

Read-only

Conversations

Configure multiple conversations with messages and assertions for this benchmark

(1)

Conversation 1

Title

Messages

Message 1

Assertion Prompt *

Override Date & Time (Optional)

Override User (Optional)