Sailboxes
Sailboxes are kernel-isolated Linux VMs with persistent state and no runtime limit, built as the cloud environment for long-horizon AI agents and driven from a CLI or SDKs.
Find alternatives to Sailboxes
What it does
- Terminal access
- Agent orchestration
Features 5
- Pause and resumeA sailbox can be paused and resumed with memory and processes intact.
- Automatic checkpointingSailbox state is snapshotted to S3 on a recurring interval.
- Per-sailbox VM isolationEach agent workload runs in its own Linux VM with a separate kernel.
- Credential injection and vaultsSecrets are stored in vaults and injected so agents authenticate without seeing API keys.
- Outbound network controlsOutbound connections can be restricted to approved destinations, blocked by host, or disabled.
Sources
- Pricing
-
Home page plan tiers read 2026-11-16: Self-Serve "Free" with "200 reserved vCPUs across active Sailboxes", Pro "$250" "/month" with "10,000 reserved vCPUs across active Sailboxes", and Enterprise "Custom"; Sailbox compute itself is still metered by the second.
- Description
-
Official documentation · 11 Nov 2026
Product page: "Sailboxes are persistent Linux VMs specifically designed to be the cloud environment for long-horizon AI agents ."
- What it does
-
Official documentation · 11 Nov 2026
Product page: "Sailboxes are persistent Linux VMs specifically designed to be the cloud environment for long-horizon AI agents ."
- Sold within
-
Sold on its own with published plans.
- Name
-
Official documentation · 11 Nov 2026
Name as published on the product page.
- URL
-
Official documentation · 11 Nov 2026
Product page, read 2026-11-11.
- Deployment
-
Official documentation · 11 Nov 2026
Product page: "Run Sailboxes with ease using the ergonomic CLI / TUI or SDKs in Python, TypeScript and Rust."
- Interfaces
-
Official documentation · 11 Nov 2026
Product page: "Run Sailboxes with ease using the ergonomic CLI / TUI or SDKs in Python, TypeScript and Rust."
Also from Sail Research
-
Sail Inference Model API
Pay-per-token API serving open models such as Kimi, GLM, DeepSeek, Gemma and Qwen through OpenAI- and Anthropic-compatible endpoints, with Default, Balanced and Flex service tiers that trade latency for a lower price per token.
Open Sail Research in the directory
Something wrong here? Send a correction — quote this product id: sail-research-sailboxes.