James Routley

Sub-millisecond VM sandboxes for AI agents via copy-on-write forking

curl -X POST https://api.zeroboot.dev/v1/exec \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer zb_demo_hn2026' \
  -d '{"code":"import numpy as np; print(np.random.rand(3))"}'

Metric	Zeroboot	E2B	microsandbox	Daytona
Spawn latency p50	0.79ms	~150ms	~200ms	~27ms
Spawn latency p99	1.74ms	~300ms	~400ms	~90ms
Memory per sandbox	~265KB	~128MB	~50MB	~50MB
Fork + exec (Python)	~8ms	-	-	-
1000 concurrent forks	815ms	-	-	-

Each sandbox is a real KVM virtual machine with hardware-enforced memory isolation.

  Firecracker snapshot ──► mmap(MAP_PRIVATE) ──► KVM VM + restored CPU state
                              (copy-on-write)         (~0.8ms)

Template (one-time): Firecracker boots a VM, pre-loads your runtime, and snapshots memory + CPU state
Fork (~0.8ms): Creates a new KVM VM, maps snapshot memory as CoW, restores all CPU state
Isolation: Each fork is a separate KVM VM with hardware-enforced memory isolation

Python — sdk/python

from zeroboot import Sandbox
sb = Sandbox("zb_live_your_key")
result = sb.run("print(1 + 1)")

TypeScript — sdk/node

import { Sandbox } from "@zeroboot/sdk";
const result = await new Sandbox("zb_live_your_key").run("console.log(1+1)");

Working prototype. The fork primitive, benchmarks, and API are real, but not production-hardened yet. Open an issue if you're interested.

Apache-2.0

Show HN: Sub-millisecond VM sandboxes using CoW memory forking