cf-orbs

Benchmarks

cf-orbs against Amp-hosted orbs, and the Docker, E2B and Vercel Sandbox providers from Amp's examples. Every provider got the same prompt on the same project. The times count orb time only, and lower is better.

Every run counted, failures included, no best-of. Each cell is the median (nearest rank). cf-orbs ran at 4 vCPU / 12 GiB and Amp-hosted at a1.small (2 CPU, 4 GiB). Each column shows its size under its name.

    Every row, its definition and its source run are in docs/benchmarks.md.