Agent Sandbox on GKE (Google Cloud)
Google Kubernetes Engine's managed path for the open-source Agent Sandbox, running AI-generated code in gVisor-isolated pods with warm pools and a Python client. For teams on Google Cloud.
- vCPU
- not published
- Memory
- not published
- 2 vCPU + 4 GiB
- no comparable rate
- Isolation
- gVisor
- Pause / resume
- Not published
- Max session
- not published
Visit Agent Sandbox on GKE (Google Cloud) Compare side by side Request a quote Checked
Cheapest providers with comparable rates
All providers against E2B| Provider | 2 vCPU + 4 GiB$/hr | vs E2B | Relative cost | Isolation |
|---|---|---|---|---|
| Alibaba Cloud Agent Sandbox | $0.055 (source: Alibaba Cloud Agent Sandbox billing overview) | 0.33x | VM | |
| Northflank Sandboxes | $0.067 (source: Northflank Pricing) | 0.40x | Cloud Hypervisor microVM | |
| Prime Intellect Sandboxes | $0.090 (source: Prime Intellect Sandboxes docs) | 0.54x | microVM | |
| Freestyle | $0.132 (source: Freestyle Pricing) | 0.80x | VM | |
| Beam | $0.140 (source: Beam Pricing) | 0.85x | — |
Every field
Source on each value- Isolation Isolation technology the vendor states for each sandbox.
- gVisor (source: Isolate AI code execution with Agent Sandbox on GKE)
- Other isolation options Alternative runtimes the vendor offers besides the default (e.g. VM runtime, gVisor for GPU).
- Not published
- Deployment Managed cloud service, self-hosted software, or both.
- managed (source: Isolate AI code execution with Agent Sandbox on GKE)
- Billing basis What the meter counts: wall-clock allocated resources, active CPU only, requests/plan quota, session-hours, or free software.
- Not published
- vCPU price $/vCPU-hr Published on-demand CPU rate normalised to USD per vCPU-hour (per-second rates multiplied by 3,600; per-core rates divided by vCPUs per core where the vendor defines it). Lowest public tier, default region.
- Not published
- Memory price $/GiB-hr Published memory rate normalised to USD per GiB-hour (GB treated as GiB where the vendor does not distinguish). 0 means the vendor states memory is free.
- Not published
- Sandbox from $/sandbox-hr Smallest published per-sandbox hourly price, where the vendor publishes one.
- Not published
- Paid plan from $/mo Cheapest paid platform plan fee (before usage).
- Not published
- Free tier A free plan, free monthly allowance or sign-up credit exists.
- Not published
- Free tier detail What the free tier or credit includes, as published.
- Not published
- Cold start ms Vendor-published create/boot time in milliseconds. Only filled when the vendor publishes a number; see startup_claim for context (percentile, conditions).
- Not published
- Resume time ms Vendor-published resume-from-pause/standby/snapshot time in milliseconds.
- Not published
- Startup claim The vendor's startup/resume wording with conditions, quoted or closely paraphrased.
- Warm pools let new sandboxes be created "in under a second" (source: Isolate AI code execution with Agent Sandbox on GKE)
- Max session length Longest single sandbox session by plan, as published.
- Not published
- Persistence / snapshots Filesystem state can be kept between runs (snapshots, persistent disks, volumes).
- Yes (source: Isolate AI code execution with Agent Sandbox on GKE)
- Pause / resume A sandbox can be paused, suspended, hibernated or put in standby and resumed later.
- Not published
- Memory state kept Pause/hibernate preserves running memory state, not only the filesystem.
- Not published
- Fork / branch A running or snapshotted sandbox can be cloned into new sandboxes.
- Not published
- GPU GPU sandboxes are offered.
- Not published
- Self-host Can run on your own hardware or cluster (open source or licensed).
- Not published
- BYOC / VPC Vendor-managed deployment inside your own cloud account.
- Not published
- Open source Core sandbox runtime is published under an open-source licence.
- Yes (source: Isolate AI code execution with Agent Sandbox on GKE)
- Licence Open-source licence of the runtime, if any.
- Apache-2.0 (source: Isolate AI code execution with Agent Sandbox on GKE)
- SDKs / languages Official SDK languages (or supported execution languages where noted).
- Python (k8s-agent-sandbox) (source: Isolate AI code execution with Agent Sandbox on GKE)
- Regions Published regions or region count.
- Not published
- Max vCPU per sandbox vCPU Largest published vCPU count per sandbox on any self-serve or enterprise plan.
- Not published
- Max memory per sandbox GiB Largest published memory per sandbox.
- Not published
- Concurrency Published concurrent sandbox limits by plan.
- Not published
- SOC 2 SOC 2 status as stated by the vendor.
- Not published
- HIPAA Vendor offers HIPAA-compliant usage/BAA (on any plan).
- Not published
- Pricing as published
- No extra charge for Agent Sandbox; standard GKE compute pricing applies. Pod snapshots referenced in docs. Checked 2026-11-09.
Sources
- Isolate AI code execution with Agent Sandbox on GKE accessed 2026-11-09
- kubernetes-sigs/agent-sandbox (GitHub) accessed 2026-11-09
Get pricing from Agent Sandbox on GKE (Google Cloud) and two close alternatives
Tell us what you need once. We send it to matching providers and they reply to you directly.