Future AGI home page
Search...
⌘K
Introduction
What is Future AGI?
Evaluation
Overview
Quickstart
Future AGI Models
Concept
How To
Eval Definition
Knowledge Base
Overview
Concept
How To
Dataset
Overview
Concept
Adding Dataset
Create Dynamic Column
Add Annotations
Change Column Type
Create Static Column
Create Synthetic Data
Add Evaluations through SDK
Experimentation
Prototype
Overview
Quickstart
Evals for Prototype
Choose Winner
Observe
Overview
Quickstart
How to run evals?
Sessions
Alerts and Monitors
Tracing
Overview
Concept
Instrumentation ( Auto )
Manual Tracing
Optimization
Overview
Concept
How To
Prompt Workbench
Overview
Concept
How To
Protect
Overview
Concept
How to Use Future AGI Protect
MCP
MCP Server
Admin & Settings
Administration Panel
FAQs
Frequently Asked Questions (FAQ)
Community
Sign Up
Future AGI home page
Search...
⌘K
Community
Sign Up
Sign Up
Search...
Navigation
Prototype
Choose Winner
Documentation
Cookbooks
Release Notes
SDK Reference
Documentation
Cookbooks
Release Notes
SDK Reference
Prototype
Choose Winner
Choose the best performing prototype version based on key parameters such as evaluation scores, cost efficiency, latency etc.
How to choose the winner?
Go to the
Prototype
dashboard.
Click on the
Choose Winner
button.
Adjust the sliders for each metric indicating its importance on a scale from
0
(not important) to
10
(very important).
Based on the values you set for each metric, all the prototype versions are ranked.
The version with the highest overall score, considering all metrics, is selected as the winner.
Was this page helpful?
Yes
No
Previous
Overview
Future AGI's Observability platform delivers enterprise-grade monitoring and evaluation for large language models (LLMs) in production. Our solution provides deep visibility into LLM application performance through advanced telemetry data tracing and sophisticated evaluation metrics.
Next
On this page
How to choose the winner?
Assistant
Responses are generated using AI and may contain mistakes.