Simulate voice and chat agents in a rich environment, buy Future AGI on Google Cloud Marketplace, and outbound Retell voice simulation

Simulate voice and chat agents in a rich environment

You can now build a full simulation environment for your hosted agents, or any voice or chat agent, by pointing Future AGI at the agent’s code, either by sharing a GitHub repository or uploading the codebase. Future AGI generates varied scenarios that exercise the agent with real tool calls rather than stubbed responses, adds deterministic evals automatically, and applies an agent-as-judge, so you get a much deeper read on how the agent actually behaves.

What you get

A simulation environment built from your agent’s own code. Future AGI reads the codebase, generates scenarios that vary across what the agent can face, and runs them against the agent with its real tools. Every run gets deterministic evals added automatically and an agent-as-judge pass on top, so the results reflect real behavior rather than canned responses.

How to turn it on

In Simulate, open Environment and click Create environment. Point it at the agent’s code by sharing a GitHub repository or uploading a folder, choose how many scenarios to run, and click Run end to end. Review each scenario’s outcome in the run timeline.

Why this matters

Simulating against stubbed tools tells you how an agent talks, not how it acts. Running scenarios with real tool calls, deterministic checks, and a judge on top shows you what the agent actually does before it reaches users.

Who this is for

Teams building hosted agents, and anyone with a voice or chat agent whose code they can share.

Buy Future AGI on Google Cloud Marketplace

Future AGI is now available on Google Cloud Marketplace. Subscribe from the Google Cloud console and Google bills you directly, so the spend runs through your Google Cloud billing account and can count toward your Google Cloud commitment. Sign up straight from the listing, including with Continue with Google, and your usage is metered and billed through Google. Plan changes and cancellations you make in the Cloud console carry through to your Future AGI organization automatically. Available on the Pay-as-you-go, Scale, and Enterprise plans.

What you get

• Buy through the account you already have. Find Future AGI on Google Cloud Marketplace and subscribe to a plan from the console. No separate procurement process, and no card on file with us.
• Spend that can count toward your commitment. Google is the merchant of record, so Future AGI appears on your Google Cloud invoice and can draw down your existing Google Cloud commitment.
• One-step sign-up. After you subscribe, the listing takes you straight into Future AGI, where one step sets up your account. Continue with Google lands you in that same organization.
• Usage billed through Google. Future AGI reports your usage to Google, so metered charges show up in Google Cloud billing alongside the rest of your cloud spend.
• The console stays in charge. Change plans in the Cloud console and your Future AGI organization follows. Cancel there and your organization moves to the free plan, with your data intact.

How to get it

Open the Future AGI listing on Google Cloud Marketplace, pick a plan, and subscribe. The listing then takes you straight into Future AGI to finish sign-up.

Who this is for

Teams on Google Cloud, especially enterprises that want to procure through an existing Google Cloud commitment rather than a new vendor agreement.

Simulate outbound Retell voice agents and enable observability for them

You can now simulate an outbound Retell voice agent through the Agent Learning Kit and enable observability for it in the same flow. Your agent places real outbound calls, one run covers every scenario, and transcripts, recordings, and evals land in the platform exactly like a native simulation, while the agent’s calls show up in Observe without a separate setup.

Improvements

Observe

Retell voice calls captured by polling. Retell voice calls are now captured by polling Retell directly instead of relying on inbound webhooks, so calls are ingested reliably even when a webhook is missed, and high call volumes are covered page by page without gaps.

Open a trace from a fired alert with its context. Following View Trace from a fired alert now carries the alert’s own time window and filters into the trace view, so you land on the traces the alert was about instead of an unscoped list.

Attributes search, project sharing, and eval mapping in Observe. The Attributes tab in the trace and voice drawers now filters as you type, project sharing works as expected, and the eval-task variable mapping screen is smoother to work through.

Dashboard metric selection persists on save. Unselecting and reordering the metrics on a dashboard widget and saving now sticks, so reopening the editor shows the selection you made.

Eval charts honor Dataset and Eval Source filters. Dashboard widgets built on an eval metric now apply the Dataset and Eval Source filters to the query, so you can build a per-environment or per-source eval trend from a single eval template.

Eval alert thresholds show their unit. An eval alert threshold is now labeled as a 0-to-1 fraction, so the value you enter lines up with the metric you are alerting on.

Evaluation

More reliable continuous eval tasks. Eval tasks that run continuously or over large historical windows now pick up every candidate row within the run’s budget, even on dense windows, and continuous tasks stop re-running failures that have already settled.

System eval bindings keep their configuration. Editing the dataset bindings on a system evaluator keeps the rest of its configuration intact across saves.

Annotation tracing filter supports contains. Filtering annotations on tracing with the contains operator now returns the matching rows.

Agents

Falcon defaults to an agent eval. Creating an evaluation through Falcon without specifying a type now creates an agent eval by default.

Gateway and API

Gateway provider dialog hardened. The Gateway provider dialog saves cleanly when a timeout value is present, surfaces validation errors clearly instead of silently blocking Save, and shows full session IDs.

Platform

Open-source UI fixes. A round of open-source-build polish, including dark-theme eval-score colors and a working Help button.