---
title: "Agent and Model Evaluations in Gemini Enterprise Agent Platform are now GA"
description: "Agent and model evaluations in Agent Platform are now generally available, providing one engine and consistent metrics for measuring agents and models during development and against live production traffic. The service offers more than 20 pre-built metrics across quality, safety, grounding, tool use, trajectories, and reference-based tasks, plus adaptive rubrics that generate case-specific criteria and a versioned, org-wide registry for custom code-based or LLM-as-a-judge metrics. Experiments can run locally or server-side, with Cloud Storage artifacts supporting auditability and reproducibility, while case generation, user simulation, and environment simulation expand coverage and model slow or failing backends. Online monitors grade sampled production traces, chart scores over time, and issue drift alerts through configured channels. Access is available through the SDK, REST API, agents-cli, console, and ADK; model-based metrics incur standard model-call charges, while code-based and computation metrics add no additional cost."
---

# Agent and Model Evaluations in Gemini Enterprise Agent Platform are now GA

[Google](https://yomu.fyi/company/google) · Alex Martin, Dima Melnyk · Jul 31, 2026

**Type:** Announcement

## Summary

Agent and model evaluations in Agent Platform are now generally available, providing one engine and consistent metrics for measuring agents and models during development and against live production traffic. The service offers more than 20 pre-built metrics across quality, safety, grounding, tool use, trajectories, and reference-based tasks, plus adaptive rubrics that generate case-specific criteria and a versioned, org-wide registry for custom code-based or LLM-as-a-judge metrics. Experiments can run locally or server-side, with Cloud Storage artifacts supporting auditability and reproducibility, while case generation, user simulation, and environment simulation expand coverage and model slow or failing backends. Online monitors grade sampled production traces, chart scores over time, and issue drift alerts through configured channels. Access is available through the SDK, REST API, agents-cli, console, and ADK; model-based metrics incur standard model-call charges, while code-based and computation metrics add no additional cost.

## Context

Agent quality needs to be measured both against cases written during development and against the tasks an agent performs after launch. Comparing those stages with consistent scoring helps identify production drift as an agent problem rather than a measurement inconsistency.

## Approach / What changed

Agent Platform provides experiments, more than 20 pre-built metrics, adaptive rubrics, custom code-based and LLM-as-a-judge metrics, online monitors, and telemetry integrations through a shared evaluation engine and versioned organization-wide registry. Case generation, user simulation, and environment simulation extend test coverage, while server-side artifacts are stored in Cloud Storage.

## Takeaways

- Adaptive rubrics generate case-specific pass/fail tests from eval case definitions, developer instructions, and tool declarations, then grade traces with a verdict and rationale.
- Server-side experiments store every artifact in Cloud Storage, making runs auditable and reproducible; code-based and computation metrics add no additional cost.
- Environment simulation can inject mocked data, forced errors, or added latency into tool calls, allowing agents to be tested against backend failures without touching production.

**Tags:** [AI Agents](https://yomu.fyi/topic/ai-agents), [Model Evaluation](https://yomu.fyi/topic/model-evaluation), [Monitoring](https://yomu.fyi/topic/monitoring)

- Source: [Google](https://developers.googleblog.com/agent-and-model-evaluations-in-gemini-enterprise-agent-platform-are-now-ga)
- Source URL: https://developers.googleblog.com/agent-and-model-evaluations-in-gemini-enterprise-agent-platform-are-now-ga
- Ingested by Yomu: 2026-09-09T14:28:05.988Z

[Read original post](https://developers.googleblog.com/agent-and-model-evaluations-in-gemini-enterprise-agent-platform-are-now-ga)