# Agent Runtime & Sandboxes

> Where agent-generated code actually runs: isolated execution sandboxes like E2B, Daytona, and Modal Sandboxes for untrusted code, and durable execution runtimes like Temporal and Inngest that keep long-running agent workflows fault-tolerant.

5 tools in 2 subcategories. Web page: https://ailandscape.org/category/agent-runtime-sandboxes · Each tool also has a markdown version at /tool/{slug}.md

## Code Execution Sandboxes

- [Daytona](https://ailandscape.org/tool/daytona): Secure and elastic infrastructure for running AI-generated code
- [E2B](https://ailandscape.org/tool/e2b): Isolated cloud sandboxes for running AI-generated code
- [Modal Sandboxes](https://ailandscape.org/tool/modal-sandboxes): Serverless sandboxed code execution for AI agents

## Durable Execution

- [Inngest](https://ailandscape.org/tool/inngest): Event-driven durable execution engine for AI and background workflows
- [Temporal](https://ailandscape.org/tool/temporal): Durable execution platform for long-running, fault-tolerant workflows

---

Part of [AI Landscape](https://ailandscape.org), an open map of the AI ecosystem. Index for AI assistants: https://ailandscape.org/llms.txt