A curated list of benchmarks and resources to help you pick the right sandbox for your long-running agents.
| Resource | Description |
|---|---|
| ComputeSDK Benchmarks | Head-to-head benchmark suite across sandbox providers. |
| Mert Devici's Sandbox Comparison | Detailed comparison of sandbox features and performance. |
| Nilesh's Full Benchmark Report | Comprehensive benchmark report by Nilesh from Baseten (Ex-CTO of Inferless). |
| The Agent Sandbox Taxonomy | Taxonomy and classification framework for agent sandboxes by George Fahmy, founder of Stakpak. |
| Ryan Vogel's Sandbox Experiments | Hands-on experimentation thread on sandbox providers. |
| Nathan Flurry's Cost Comparison | Hourly and monthly cost comparison across sandbox providers by Nathan Flurry from Rivet. |
We're working on our own take on infra for long-running agents at opencomputer.dev — we'd love all the feedback you can offer!
Know a benchmark, comparison, or resource that should be here? Open a PR or an issue — contributions are welcome!