Google launches Agent Compute Engine, giving each Gemini agent its own hardware-isolated micro VM
Current overview (not historical)
Google has launched Agent Compute Engine, a service that gives each Gemini agent its own hardware-isolated micro virtual machine protected by Titanium chips. The VMs automatically pause when idle to release resources and use shared memory to reduce costs. Google states that VMs are the primary cost factor for Agentic systems, and the service currently supports Gemini, with broader preview access planned over the coming months.
Written with AI from the reporting · updated 10/22/2026
Expand the detailed fact route and sources
Story route
A station-by-station route from the first public evidence onward; every station carries its own evidence and uncertainty.
- 1Unclassified fact stationOct 23, 02:10single sourcereport-time onlyrevised
Google Agent Compute Engine gives each Gemini agent its own
Read original source ↗ The DecoderReplay this moment in the explorerEvidence & revisions · 1 source records
- Subject:
- Action:
- announced Agent Compute Engine
- Object:
- Gemini agent
- Occurred (fact side):
- not recorded
- Published (valid side):
- Oct 23, 01:50
- First seen (known side):
- Oct 23, 02:10
Fact stationSingle source · one-sided claimRevisedReport-time onlyCopy may have been revised later; time replay is bounded. Layer labels group reading only, not causation between layers.
Uncertainty marks flag where to check the sources; they are not proven causation.
Report timeline
Explore every side through dated source reports.
- The DecoderRelated fact: single sourceFact occurrence time unverifiedSource report revisedGoogle's Agent Compute Engine provides hardware-isolated microVM for each Gemini agent
Report first recorded Oct 23, 02:10
Google announced Agent Compute Engine, a service giving each AI agent its own isolated micro virtual machine secured by Google's Titanium chip. This service, which already powers Gemini, will open for preview in the coming months. The system features automatic VM pausing when idle, shared memory to reduce running costs, and Google claims it can fit ten times as many agents on a single server, with suspend and resume taking less than 500 milliseconds in a 200,000 agent test. Oversight features include token tracking, model routing, and support for third-party identity providers like Ping Identity and SailPoint.
Read original source ↗
Event discussion
Reader opinions are hosted on GitHub Discussions; posting requires a GitHub account. Comments are not verified facts.