CodeSandbox SDK vs Agent 37
CodeSandbox SDK gives you Firecracker VMs that hibernate with their memory, wake on the next HTTP request and fork from a warm state. Agent 37 hosts one sandbox per user with agents like Hermes, Claude Code and Codex built in, on lighter gVisor sandboxes billed per minute from a wallet.
The companies and products
CodeSandbox SDK
CodeSandbox SDK is an API from CodeSandbox, now a Together AI company, that starts sandboxed development VMs on the same infrastructure as the CodeSandbox editor.
Each sandbox is a Firecracker microVM built from a template or any Dockerfile. When it goes idle it hibernates with a memory snapshot, and it resumes or forks from that snapshot in seconds. You drive it from a TypeScript SDK through terminals, commands, files and HTTPS hosts. Together's docs now call it the legacy offering and point new users to its successor, the together-sandbox SDK.
Agent 37
Agent 37 hosts AI agents, one isolated, always-on sandbox per user, behind one API.
You create an instance from a template (Hermes, OpenClaw, Claude Code, Codex, OpenCode, Pi, Grok or your own Docker image) and talk to it through a chat API, files, exec, SSH and HTTPS URLs. The default instance is a container enforced by gVisor; a VM type gives a full Linux machine with its own kernel for Docker, Kubernetes and Android emulators. It is built for companies that give each of their customers an agent.
- Where it runs
- Dedicated servers in the United States
- Instance types
- Default (shared), Performance (dedicated cores), VM
- Compliance
- SOC 2 Type I, Type II in progress
Which use cases each one fits
For each use case, Battleships prices every sandbox provider on the same workload and sorts them from the cheapest monthly bill up. “46th cheapest of 345” means 45 of the other 344 providers run that workload, with every feature it needs, for less. “With limits” means the provider runs it but misses something, so it is listed after every full match. Open a use case for its whole ranking and what each provider misses.
| Use case and its workload | CodeSandbox SDK: price rank | Agent 37: price rank | bochka: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month CodeSandbox SDK runs Docker on every VM but has no SSH; Agent 37 needs its VM type for Docker; bochka gives every sandbox Docker, systemd and SSH. | 178th of 367, with limits $1,939 a month | 2nd cheapest of 367 $551 a month | 3rd cheapest of 367 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Agent 37 is built for this: one always-on assistant per user that sleeps when idle, keeps its memory and wakes on the next message, while CodeSandbox SDK is a compromise here. | 215th of 366, with limits $2,478 a month | Cheapest of 366 $279 a month | 5th cheapest of 366 $657 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Both suit short code runs: CodeSandbox SDK wakes hibernated VMs with their memory, and Agent 37's small default sandboxes cost very little per minute. | 46th cheapest of 345 $642 a month | Cheapest of 345 $22 a month | 186th of 345, with limits $2,000 a month |
| Sandboxes for your users 300 at once · 2 vCPU / 4 GB · 30,000 sessions of 23 min a month Neither fits 300 users at once on self-serve: CodeSandbox SDK's Scale plan stops at 250 VMs and Agent 37 at 200 instances, so bochka fits best here. | 140th of 361, with limits $1,707 a month | 30th of 361, with limits $79 a month | Cheapest of 361 $209 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month CodeSandbox SDK's forks share memory, so many rollouts can branch from one running state; Agent 37 has no fork. | 348th of 360, with limits $3,861 a month | 6th of 360, with limits $363 a month | 34th of 360, with limits $2,000 a month |
| Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month | 64th cheapest of 364 $1,335 a month | 90th of 364, with limits $58 a month | 10th cheapest of 364 $200 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month Both run Docker builds, Agent 37 only on its VM type; jobs that need /dev/kvm work on Agent 37's VM type and on every bochka sandbox. | 92nd cheapest of 377 $860 a month | 2nd cheapest of 377 $84 a month | 4th cheapest of 377 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB | 310th of 368, with limits $240 a month | 3rd cheapest of 368 $12 a month | 18th cheapest of 368 $26 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Agent 37 adds a noVNC desktop through a template build, CodeSandbox SDK has no desktop, and bochka streams a 60fps desktop out of the box. | 206th of 369, with limits $691 a month | Cheapest of 369 $53 a month | 6th cheapest of 369 $200 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month | 251st of 367, with limits $443 a month | 48th of 367, with limits $29 a month | 7th cheapest of 367 $200 a month |
| Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month | 281st of 346, with limits $508 a month | Cheapest of 346 $32 a month | 6th cheapest of 346 $90 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month | 193rd of 338, with limits $7,390 a month | 36th of 338, with limits $648 a month | 71st of 338, with limits $1,800 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month Agent 37 is a strong fit for many cheap parallel attempts; bochka can also fork a running VM to try many paths from the same point. | 240th of 360, with limits $326 a month | Cheapest of 360 $43 a month | 2nd cheapest of 360 $200 a month |
| GPU eval jobs 10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month | Can't run it | Can't run it | Can't run it |
| macOS agents 5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month | Can't run it | Can't run it | Can't run it |
Key features
| Feature | CodeSandbox SDK | Agent 37 | bochka |
|---|---|---|---|
| Isolation | Firecracker microVM with its own kernel | gVisor sandbox by default; full VM as an opt-in instance type | Every sandbox is a full VM with its own kernel, on dedicated AMD Ryzen 9 9950X hosts |
| Docker inside | Docker and root on every VM, built from any Dockerfile | Only on the VM type, at 2.5x the default compute rate | Docker, systemd and root in every sandbox |
| Nested virtualization | Not documented | /dev/kvm on the VM type, with a ready Android template | /dev/kvm in every sandbox, Android emulators work |
| Shell access | Terminals, commands and files through the SDK; no SSH | SSH through a WebSocket tunnel, plus exec and files | SSH, plus exec and files over the API |
| Memory across sleep | Hibernates with a memory snapshot after an idle timeout of up to one day | Opt-in auto-sleep (default 15 minutes) checkpoints memory | Disk saved every minute and on stop; a resume reboots the machine |
| Wake on request | Yes, an HTTP request to a host wakes it | Yes, wakes on the next request | No; a stopped sandbox is started by API |
| Fork | Fork a hibernated sandbox from its snapshot, or a running one sharing its memory | No fork; daily backups plus one on-demand backup, restored in place only | Fork a running sandbox with its disk; named snapshots as templates |
| Biggest machine | 16 cores / 32 GB on Scale, up to 64 cores / 128 GB on Enterprise | Shapes up to 16 vCPU / 32 GB | Up to 16 vCPU / 32 GB on the $100 plan and above |
| Custom images | Templates from any Dockerfile | Your own linux/amd64 Docker image as a template | Templates from named snapshots, no Dockerfile builds |
| Built-in agents | None; run your own agent through the command API | Hermes, OpenClaw, Claude Code, Codex, Grok, OpenCode, Pi, with a chat API | Claude Code, Codex, pi, OpenCode and others, with a prompt, steer and events API |
| Desktop and browser | No desktop; a headless Chromium template | noVNC desktop through a template build or the Ubuntu VM template | 60fps desktop with Chromium and computer-use tools in every sandbox |
| Keys kept out of the sandbox | Secrets go in as environment variables | Managed LLM, search and Composio via a per-instance token; your own keys go in as env vars | Credentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out |
| Inbound access | HTTPS host per port, public by default, signed tokens for private hosts; no raw TCP or public IP | HTTPS URL per port, signed URLs, custom domains; no public IP | HTTPS URLs for up to 50 ports, token-protected by default; raw TCP; a dedicated IPv6 or IPv4 |
| Regions | A country hint steers scheduling, for example to Europe, but placement is not guaranteed | Dedicated servers in the United States | EU only (Germany, Finland, France), with EU data residency |
Every feature Battleships tracks (43)
| Feature | CodeSandbox SDK | Agent 37 | bochka |
|---|---|---|---|
| Full VM (own kernel) | Yes | No | Yes |
| Docker inside | Yes | No | Yes |
| Nested virtualization | No | No | Yes |
| Root | Yes | Yes | Yes |
| Systemd | Not published | Not published | Yes |
| Desktop GUI | No | Yes | Yes |
| Browser | No | Yes | Yes |
| Computer-use API | No | Yes | Yes |
| SSH | No | Yes | Yes |
| Persistent disk | Yes | Yes | Yes |
| Pause/resume | Yes | Yes | Yes |
| Automatic snapshots | Yes | Yes | Yes |
| Fork/clone | Yes | Yes | Yes |
| Live fork (no pause) | Yes | Yes | Yes |
| Memory fork | Yes | No | No |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | Yes | Yes | No |
| Wake on request | Yes | Yes | No |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | No | Not published | Yes |
| Public IPv4 | No | No | No |
| Egress allowlist | No | No | No |
| Secret proxy | No | Yes | Yes |
| Secret proxy for any API | No | No | Yes |
| Preinstalled agents | No | Yes | Yes |
| Agent harness API | No | Yes | Yes |
| MCP server | No | No | No |
| GPU | No | No | No |
| Windows | No | No | No |
| Arm64 | No | No | No |
| Self-hosting | No | No | No |
| BYOC | No | No | No |
| Open source | No | No | No |
| Scoped API keys | Not published | Not published | Yes |
| Spend limits | Not published | Yes | Yes |
| Usage per sandbox (API) | No | Yes | Yes |
| Audit logs | Not published | Not published | No |
| SSO | Yes | No | No |
| SOC 2 | Yes | Yes | No |
| HIPAA | No | No | No |
| EU data residency | No | No | Yes |
| Zero data retention | Not published | Not published | Yes |
Performance
Neither CodeSandbox SDK nor Agent 37 has a comparable CPU measurement. bochka did 26.69 runs per second on the StarSling hpc-sandbox-benchmarks Node.js loop at 4 vCPU / 8 GB, on dedicated AMD Ryzen 9 9950X hosts. Agent 37 says its default instances share host resources and sells a Performance type on dedicated cores.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. CodeSandbox SDK and Agent 37 have no published run in this suite yet, so there is no like-for-like number to compare.
Pricing in five scenarios
Each bill is priced from the providers' public price lists by the open Battleships engine, with the cheapest plan that runs the scenario.
| Scenario | CodeSandbox SDK, a month | Agent 37, a month | bochka, a month |
|---|---|---|---|
| One coding agent, workdays 4 vCPU / 8 GB, 50 GB disk, 8 hours a day, 22 days a month, files kept between days | $193 · Scale | $6 · Wallet (any top-up) | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $341 · Scale | $12 · Wallet (any top-up) | $26 · $20/mo |
| 50 agents in parallel 50 machines of 4 vCPU / 8 GB at once, 8-hour sessions, 1,100 sessions a month, Docker inside | $2,500 · Scale | $329 · Wallet ($100 lifetime top-ups) | $317 · $20/mo × 3 seats |
| Sandboxes for your users 2 vCPU / 4 GB, 30,000 sessions of 10 minutes a month, up to 300 at once | $815 · Scale · with limits | $33 · Wallet ($250 lifetime top-ups) · with limits | $200 · $100/mo × 2 seats |
| 1,000 short jobs at once 2 vCPU / 4 GB, 1,000 at once, 3 minutes each, 200,000 jobs a month | $1,632 · Scale · with limits | $73 · Wallet ($250 lifetime top-ups) · with limits | $2,000 · $2000/mo · with limits |
CodeSandbox SDK: You buy VM credits and each size burns a fixed number per hour, billed per started minute while the VM runs, including idle time before it hibernates. Scale adds a $170 monthly platform fee that is not usage credit, and a hibernated VM costs nothing.
Agent 37: A prepaid wallet with no plan fee, metered per minute: $0.80 per vCPU and $0.70 per GB of RAM per month while running, and $0.09 per GB of disk per month even while asleep. Performance compute costs 4x and VM compute 2.5x the default rate.
Agent 37: Plans start at $20 a month, and all of it is usage credit. You pay per second while a sandbox runs, and a stopped sandbox costs nothing and keeps its disk.
Limits in those scenarios
CodeSandbox SDK's Scale plan allows 250 VMs at once, and Agent 37 allows 200 instances on its top self-serve tier, counting sleeping and stopped ones, so neither fits 300 user sandboxes or 1,000 jobs at once. Both cap disk at 20 GB on a 4 vCPU / 8 GB machine, and neither publishes an egress price. The 50-agent scenario needs Docker, so Agent 37 is priced on its VM type. bochka fits the user-sandboxes scenario, but its start limits queue the 1,000-job burst even on the $2,000 plan.
| Scenario | CodeSandbox SDK | Agent 37 | bochka |
|---|---|---|---|
| One coding agent, workdays |
|
| None that bite |
| One agent that never stops |
|
| None that bite |
| 50 agents in parallel |
|
| None that bite |
| Sandboxes for your users |
|
| None that bite |
| 1,000 short jobs at once |
|
|
|
Which one should you use?
Use CodeSandbox SDK when
your sandboxes sleep between visits and should wake on the next HTTP request with their memory intact, or when you fork many runs from one warm state, and you are fine building on what Together now calls its legacy SDK.
- Hibernation with memory and wake on request. Good for previews and code runs used in bursts: an HTTP request wakes a hibernated VM with its processes still running.
- Forks that share memory. Good for branching many runs from one warm state, such as rollouts or test variants, without booting each copy again.
- Docker everywhere, bigger machines on Enterprise. Every VM runs Docker from any Dockerfile, and Enterprise teams can get single machines up to 64 cores and 128 GB.
Use Agent 37 when
you host one mostly idle assistant per customer in the US, from a ready agent template, and want a small per-minute bill with memory kept while it sleeps.
- Built for one agent per customer. Pick a template like Hermes, OpenClaw, Claude Code or Codex and talk to each user's instance through a chat API.
- Small bills for mostly idle assistants. A prepaid wallet billed per minute with no plan fee, and idle instances can sleep with their memory kept.
- Your Docker image, dedicated cores, US servers. Build templates from your own Docker image, pick dedicated cores on the Performance type, and run close to US users.
Where bochka fits
Both are good at sandboxes that sleep and wake. If your agents instead work for hours on installs, builds, Docker, emulators and a browser, bochka gives every sandbox a full VM with SSH, a desktop and coding agents set up, and you can fork it while it runs.
A full machine by default
CodeSandbox SDK has no SSH and no documented /dev/kvm, so emulators cannot start. Agent 37's default gVisor type cannot run Docker; its VM type can, at 2.5x the compute rate and with no init system.
A desktop for computer use
CodeSandbox SDK has no desktop, only a headless Chromium template you build on. Agent 37 adds a noVNC desktop through a template build. bochka streams a 60fps desktop with Chromium in every sandbox.
Keys outside the sandbox, for any API
On CodeSandbox SDK, and on Agent 37 for your own keys, secrets sit as env vars inside a machine the agent controls, where one prompt injection can leak them. bochka's proxy keeps them outside.
Use bochka when your agents do real work for hours: installs, builds, Docker, emulators and a browser on a full VM, with forks of running machines, servers others can reach and keys kept outside for any API. Compare it one to one: bochka vs CodeSandbox SDK and bochka vs Agent 37.
Questions people ask
Is CodeSandbox SDK cheaper than Agent 37?
No. Agent 37 comes out lower in every scenario on this page. One coding agent that never stops is $12 a month on Agent 37 against $341 on CodeSandbox SDK, and 50 agents in parallel are $329 against $2,500.
Which is faster, CodeSandbox SDK or Agent 37?
Neither has a comparable CPU measurement, so there is no fair speed answer. Agent 37 sells a Performance type on dedicated cores for steady speed. bochka did 26.69 runs per second on the same benchmark at 4 vCPU / 8 GB.
Can CodeSandbox SDK and Agent 37 run Docker?
CodeSandbox SDK runs Docker on every VM and builds from any Dockerfile. Agent 37 runs Docker only on its VM type, which bills compute at 2.5x; the default gVisor type cannot run a Docker daemon. bochka runs Docker in every sandbox.
Do both keep memory when a sandbox sleeps?
Yes. CodeSandbox SDK hibernates with a memory snapshot, and Agent 37 checkpoints memory on auto-sleep, and both wake on the next request. bochka keeps the disk but reboots the machine on resume.
Is CodeSandbox SDK still maintained?
CodeSandbox is now part of Together AI. Together's docs call the CodeSandbox-based SDK deprecated and name together-sandbox as its successor, while self-serve sign-up still runs through CodeSandbox.
Where do they run?
Agent 37 runs on dedicated servers in the United States. CodeSandbox SDK takes a country hint, for example to a cluster in Europe, but does not guarantee placement. bochka runs only in the EU, with EU data residency.
Try bochka
Install the CLI, then run bochka new.
Product preview
More comparisons
- bochka vs CodeSandbox SDK
- bochka vs Agent 37
- E2B vs CodeSandbox SDK
- Daytona vs CodeSandbox SDK
- Modal vs CodeSandbox SDK
- Vercel Sandbox vs CodeSandbox SDK
- Cloudflare Sandbox vs CodeSandbox SDK
- Blaxel vs CodeSandbox SDK
- Runloop vs CodeSandbox SDK
- Tensorlake vs CodeSandbox SDK
- Morph Cloud vs CodeSandbox SDK
- CodeSandbox SDK vs Northflank