Vercel Sandbox vs Koyeb
Both run each sandbox in its own microVM. Vercel Sandbox is built for short runs of untrusted code in 19 regions, with sessions capped at 24 hours, while Koyeb runs sandboxes as services that can sleep when idle, wake on the next request, and use GPUs.
The companies and products
Vercel Sandbox
Vercel is a cloud platform for web apps and agents, and Vercel Sandbox is its product for running untrusted or agent-written code in isolated Linux microVMs.
Each sandbox is a Firecracker microVM with its own kernel, started from a Vercel image, your own OCI image, or a saved snapshot. You drive it with the JavaScript or Python SDK or the sandbox CLI: run commands, move files, open ports, and set network rules. It is aimed at code interpreters, app builders, test runs and agent tools, and Vercel's docs say it is not meant to run nonstop.
Koyeb
Koyeb is a serverless cloud for apps, inference and sandboxes, and in February 2026 it agreed to join Mistral AI.
A Koyeb sandbox is a Koyeb service: a microVM started from any Docker image and driven from Python or JavaScript SDKs that run commands, read files and expose a port. It can scale to zero when idle and wake on the next request, and it can run on CPU or GPU instance types. It is built for teams that run AI agents or untrusted code next to apps they already host on Koyeb.
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. “22nd cheapest of 345” means 21 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 | Vercel Sandbox: price rank | Koyeb: price rank | bochka: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Both are compromises for a software factory: Vercel stops each session at 24 hours, Koyeb's local disk is lost on redeploy, and neither has a desktop or /dev/kvm; bochka ranks 3rd. | 244th of 367, with limits $3,351 a month | 54th of 367, with limits $366 a month | 3rd cheapest of 367 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An assistant that lives on its machine all month hits Vercel's 24-hour session cap, and on Koyeb a deep sleep throws away the VM; bochka keeps its disk with no session cap. | 257th of 366, with limits $3,657 a month | 135th of 366, with limits $923 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 Short code bursts fit both: Koyeb ranks 4th with small instances that scale to zero, and Vercel ranks 22nd, billing CPU only while code uses it. | 22nd cheapest of 345 $216 a month | 4th cheapest of 345 $66 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 Both fit sandboxes for your own users; Koyeb ranks 6th, Vercel ranks 15th, and bochka ranks first on cost for the same work. | 15th cheapest of 361 $1,267 a month | 6th cheapest of 361 $495 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 Vercel fits many short parallel rollouts well and ranks 3rd, while Koyeb is a compromise here. | 3rd cheapest of 360 $15,761 a month | 237th of 360, with limits $1,631 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 | 69th cheapest of 364 $1,508 a month | 19th cheapest of 364 $429 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 | 107th cheapest of 377 $1,144 a month | 142nd of 377, with limits $88 a month | 4th cheapest of 377 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB A dev box that stays up all day restarts every 24 hours on Vercel and needs a volume to keep files on Koyeb; on bochka it keeps running and keeps its disk. | 276th of 368, with limits $145 a month | 217th of 368, with limits $37 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 Neither Vercel Sandbox nor Koyeb has a desktop for a computer-use agent; bochka streams a 60fps Linux desktop with Chromium and ranks 6th. | 235th of 369, with limits $952 a month | 81st of 369, with limits $123 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 | 259th of 367, with limits $486 a month | 172nd of 367, with limits $123 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 | 277th of 346, with limits $468 a month | 172nd of 346, with limits $89 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 | 234th of 338, with limits $9,960 a month | 65th of 338, with limits $1,683 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 | 283rd of 360, with limits $554 a month | 233rd of 360, with limits $299 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 Koyeb runs GPU instances from RTX 4000 cards up to H100 and H200; Vercel Sandbox and bochka have no GPUs. | Can't run it | 3rd cheapest of 98 $628 a month | 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 | Vercel Sandbox | Koyeb | bochka |
|---|---|---|---|
| Isolation | Firecracker microVM with its own kernel | MicroVM per instance | Full VM with its own kernel on dedicated AMD Ryzen 9 9950X hosts |
| Longest session | 24 hours on Pro and Enterprise, 45 minutes on Hobby; resume starts a new session | No forced cap; auto-delete timers are optional | No cap on paid plans (--no-auto-stop); default TTL 1 hour, changeable |
| Saving files | Filesystem snapshot on stop, billed per GB a month and kept 30 days after last use by default | Local disk is lost on redeploy; volumes (preview, 1 to 10 GB, two regions) persist; SDK snapshots are experimental | Disk saved every minute while running and on stop; a stopped sandbox costs nothing |
| Sleep and wake | No idle auto-stop; runs until its timeout or stop() | Scales to zero after an idle period (5 minutes minimum) and wakes on the next request; Light Sleep wakes in about 200 ms | Stop and start by API; a stopped sandbox costs nothing and keeps its disk |
| CPU speed (4 vCPU / 8 GB) | 12.96 runs/s on the Node.js benchmark | Not benchmarked; at least 0.25 dedicated vCPU per GB of memory, bursting to all cores | 26.69 runs/s on the Node.js benchmark |
| Biggest machine | 8 vCPU / 16 GB on Pro, 32 vCPU / 64 GB on Enterprise | Instance types up to 40 vCPU / 128 GB, and GPU types far larger | Up to 16 vCPU / 32 GB on the $100 plan and above |
| GPUs | No | RTX 4000 SFF Ada up to H100, H200 and multi-GPU | No |
| Nested virtualization | No | Not documented | Yes, /dev/kvm for Android emulators and VMs |
| Desktop, browser, computer use | None; headless Linux only | No desktop or browser in the sandbox SDK | 60fps desktop and Chromium, computer-use and browser-control APIs, embeddable stream |
| Coding agents | Claude Code, Codex, OpenCode and pi preinstalled; you start them as commands | Tutorials that run agents through the command API; an MCP server lets agents manage Koyeb | Seven agents preinstalled, with an API to prompt, steer and stream events |
| Inbound access | Public URLs for up to 15 ports, no built-in auth; no raw TCP or public IP | One exposed HTTPS port at a time, plus an optional TCP proxy to one port; no dedicated IP | HTTPS URLs for up to 50 ports, token-protected by default; raw TCP; dedicated IPv4 or IPv6 |
| Egress rules | Allow-all, deny-all or a domain and CIDR allowlist, changeable while running; static IPs on Enterprise | Block all outbound traffic, or allow only listed IPs and CIDRs | Open internet, no provider-side allowlist; firewall inside the VM |
| Keeping API keys out of the sandbox | Credentials brokering injects headers for any HTTPS domain you allow | Secrets are set as environment variables inside the service | Credentials proxy for model providers and any HTTPS API; switch a credential to another account when credits run out |
| Regions | 19 regions across the US, Europe, Asia, South America, Africa and Australia | Frankfurt, Paris, Washington D.C., AWS US East, Singapore, Tokyo, plus San Francisco in preview | EU only: Germany, Finland, France |
Every feature Battleships tracks (43)
| Feature | Vercel Sandbox | Koyeb | bochka |
|---|---|---|---|
| Full VM (own kernel) | Yes | Yes | Yes |
| Docker inside | Yes | Yes | Yes |
| Nested virtualization | No | No | Yes |
| Root | Yes | Not published | Yes |
| Systemd | Not published | Not published | Yes |
| Desktop GUI | No | No | Yes |
| Browser | No | No | Yes |
| Computer-use API | No | No | Yes |
| SSH | No | No | 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 | Not published | Yes |
| Memory fork | No | Yes | No |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | No | Yes | No |
| Wake on request | No | Yes | No |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | No | Yes | Yes |
| Public IPv4 | No | No | No |
| Egress allowlist | Yes | Yes | No |
| Secret proxy | Yes | No | Yes |
| Secret proxy for any API | Yes | No | Yes |
| Preinstalled agents | Yes | Yes | Yes |
| Agent harness API | No | No | Yes |
| MCP server | Yes | Yes | No |
| GPU | No | Yes | No |
| Windows | No | No | No |
| Arm64 | No | No | No |
| Self-hosting | No | Yes | No |
| BYOC | No | Yes | No |
| Open source | No | No | No |
| Scoped API keys | Not published | Not published | Yes |
| Spend limits | Yes | No | Yes |
| Usage per sandbox (API) | No | No | Yes |
| Audit logs | Not published | Yes | No |
| SSO | Yes | Yes | No |
| SOC 2 | Yes | Yes | No |
| HIPAA | Yes | No | No |
| EU data residency | Yes | Yes | Yes |
| Zero data retention | Not published | Not published | Yes |
Performance
On the hpc-sandbox-benchmarks Node.js build-and-test loop at 4 vCPU / 8 GB, bochka did 26.69 runs per second and Vercel Sandbox did 12.96, so bochka finished about twice as many runs in the same time. There is no comparable measurement for Koyeb: it was not part of that run. Koyeb publishes a CPU share of at least 0.25 dedicated vCPU per GB of memory, bursting to all cores, so its speed depends on how busy the host is.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Koyeb has 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 | Vercel Sandbox, a month | Koyeb, 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 | $57 · Pro | $29 · Pro | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $236 · Pro · with limits | $57 · Pro | $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 | $2,884 · Pro | $475 · Pro | $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 | $828 · Pro | $329 · Scale | $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 | $3,152 · Pro | $485 · Scale · with limits | $2,000 · $2000/mo · with limits |
Vercel Sandbox: Vercel charges for CPU only while your code uses it, but charges for memory for the whole time the sandbox runs, idle time included, plus a small fee per sandbox created and a monthly fee per GB of snapshots. The $20 Pro fee is usage credit shared with your other Vercel usage.
Koyeb: Koyeb bills each instance per second from the moment it starts, at a fixed price per instance type, and Deep Sleep costs nothing. Pro costs $29 a month with $10 of usage included and Scale $299 a month with $100 included; the rest of the fee is not credit.
Koyeb: bochka charges a flat price per size, per second, only while the machine runs, and a stopped sandbox costs nothing. Plans start at $20 a month, and all of the fee is usage credit.
Limits in those scenarios
For one agent that never stops, Vercel ends each session at 24 hours; the sandbox resumes from its snapshot, but running processes start over. Koyeb's priced shape, eco-xlarge, has a 20 GiB disk with no published price for more. For 1,000 short jobs at once, Vercel fits, though its vCPU allocation starts at 150 a minute and climbs with steady use, while Koyeb's eco instances guarantee only half of each vCPU and throttle above that. bochka does not fit that burst either: its $2,000 plan allows 600 starts an hour and queues bursts above 90 a minute.
| Scenario | Vercel Sandbox | Koyeb | bochka |
|---|---|---|---|
| One coding agent, workdays | None that bite |
| None that bite |
| One agent that never stops |
|
| None that bite |
| 50 agents in parallel | None that bite | None that bite | None that bite |
| Sandboxes for your users | None that bite | None that bite | None that bite |
| 1,000 short jobs at once | None that bite |
|
|
Which one should you use?
Use Vercel Sandbox when
you run many short bursts of untrusted code, already build on Vercel, or need to run close to users in many regions.
- Regions all over the world. Useful if your users are spread out: Vercel Sandbox runs in 19 regions, from Virginia to Tokyo, Sydney and Cape Town.
- A firewall you control from outside. A running sandbox can switch to deny-all, or to a list of allowed domains, without a restart.
- Images from your Dockerfile. Useful if you already deploy on Vercel: build a Dockerfile into an image and boot sandboxes straight into it.
Use Koyeb when
your sandboxes need GPUs, sleep most of the time and must wake on a request, or run for longer than a day.
- GPUs in the same platform. Useful when an agent or eval needs a GPU: Koyeb runs instances from RTX 4000 cards up to H100 and H200.
- Sleep when idle, wake on request. Good for sandboxes that sit idle between requests: Koyeb scales them to zero and wakes them on the next one.
- No session cap. A Koyeb sandbox keeps running until you delete it, so long jobs are not cut off at 24 hours.
Where bochka fits
If your agents work for hours on one machine, both Vercel Sandbox and Koyeb leave gaps: Vercel stops each session at 24 hours, and Koyeb can lose its local disk. bochka gives each agent a full VM that keeps its disk, with a desktop, Docker, /dev/kvm and an API to drive the agents inside from outside. On the same CPU benchmark it did about twice the runs of Vercel Sandbox.
A disk that stays, saved every minute
Vercel snapshots only on stop and restarts every process after 24 hours. Koyeb loses the local disk on redeploy and a deep sleep discards the VM, so you add a preview volume of 1 to 10 GB or experimental snapshots.
A desktop, a browser and /dev/kvm
Neither Vercel Sandbox nor Koyeb has a desktop or browser, so an agent cannot open the app it built unless you build that screen yourself. Vercel has no nested virtualization and Koyeb does not document it, so Android emulators run elsewhere.
Agents, keys and ports handled for you
On both, coding agents run as plain commands and you write the harness to prompt and steer them. Vercel exposes 15 ports with no built-in auth, Koyeb one port at a time with keys as readable environment variables; bochka gives 50 token-protected HTTPS ports and a credentials proxy.
Use bochka when your agents work for hours on a machine that keeps its disk: builds, Docker, emulators, a desktop browser and preview servers, with an API to drive the agents inside from outside. Compare it one to one: bochka vs Vercel Sandbox and bochka vs Koyeb.
Questions people ask
Is Vercel Sandbox cheaper than Koyeb?
No, Koyeb costs less in all five of our scenarios. One workday agent costs $29 a month on Koyeb against $57 on Vercel, and 50 parallel agents cost $475 against $2,884. For 1,000 short jobs at once Koyeb is $485 but its eco CPU throttles, while Vercel fits at $3,152 a month.
Which is faster, Vercel Sandbox or Koyeb?
Only Vercel Sandbox was measured: 12.96 runs per second on the Node.js benchmark at 4 vCPU / 8 GB, against 26.69 for bochka. Koyeb was not part of that run, and it guarantees at least 0.25 dedicated vCPU per GB of memory.
Can Vercel Sandbox or Koyeb run 24/7?
Koyeb can: a sandbox keeps running until you delete it, though its local disk is lost on redeploy. Vercel Sandbox sessions stop after 24 hours on Pro and Enterprise and resume from a snapshot with processes started fresh.
Can they run Docker and Android emulators?
Vercel Sandbox runs Docker inside with sudo, and Koyeb sandboxes start from any Docker image. Vercel has no nested virtualization and Koyeb does not document it, so Android emulators need another machine. bochka runs Docker with root and systemd and exposes /dev/kvm.
Which one has GPUs?
Koyeb does, from an RTX 4000 SFF Ada up to H100 and H200, billed per second. Vercel Sandbox and bochka are CPU only.
Is bochka SOC 2 compliant?
SOC 2 is in progress. We're open about where we stand: live status at status.bochka.xyz and our data retention policy at docs.bochka.xyz/data-retention.
Try bochka
Install the CLI, then run bochka new.
Product preview
More comparisons
- bochka vs Vercel Sandbox
- bochka vs Koyeb
- E2B vs Vercel Sandbox
- Daytona vs Vercel Sandbox
- Modal vs Vercel Sandbox
- Vercel Sandbox vs Cloudflare Sandbox
- Vercel Sandbox vs Blaxel
- Vercel Sandbox vs Runloop
- Vercel Sandbox vs Tensorlake
- Vercel Sandbox vs Morph Cloud
- Vercel Sandbox vs CodeSandbox SDK
- Vercel Sandbox vs Northflank