Modal vs Upstash Box
Modal runs sandboxes on a large AI compute platform with GPUs and regions around the world. Each sandbox lasts at most 24 hours, and your code keeps its disk with a filesystem snapshot. Upstash Box gives each agent a light container in AWS us-east-1 that pauses when idle, keeps its files and wakes on the next request.
The companies and products
Modal
Modal is a serverless cloud for AI work: inference, training, batch jobs and sandboxed code, all driven from Python, TypeScript or Go code.
Modal Sandboxes are containers you create from code, run commands in, and then terminate. By default they run under gVisor, a user-space kernel from Google. Since October 2026 you can also pick a VM runtime with its own Linux kernel for Docker and FUSE (Battleships tracks that as a separate card, Modal VM Sandboxes). Your code can save a sandbox's disk as a filesystem snapshot and start new sandboxes from it. It suits teams that already run models on Modal and want scalable code execution next to their GPUs.
Upstash Box
Upstash Box is a cloud computer for AI agents from Upstash, the company known for serverless Redis, Vector and QStash.
Each box is a Docker container with its own filesystem, shell, git and an optional coding agent such as Claude Code, Codex, OpenCode or Cursor, which you drive from TypeScript or Python SDKs. A box pauses when it sits idle, keeps its files, and you pay for the CPU it actually uses. Upstash built it to run its own Context7 service and opened it to everyone in March 2026.
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. “15th cheapest of 336” means 14 of the other 335 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 | Modal: price rank | Upstash Box: price rank | boat: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month A software factory needs Docker and files kept between sessions. Modal's VM runtime ranks 7th here, boat 8th. Upstash (70th) and Modal's default gVisor runtime (76th) are compromises, because neither documents Docker inside. | 106th of 365, with limits $633 a month | 98th of 365, with limits $464 a month | 5th cheapest of 365 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB boat ranks 5th, Modal 48th and Upstash 106th; both rivals are compromises here. Modal ends each sandbox at 24 hours, so your code snapshots the disk and starts a new one. boat has no session cap once auto-stop is off (default 1 h). | 61st of 361, with limits $1,574 a month | 170th of 361, with limits $450 a month | 6th cheapest of 361 $657 a month |
| Code interpreter bursts 100 at once · 1 vCPU / 2 GB · 200,000 sessions of 1 min a month Both fit short code runs: Upstash Box ranks 10th and Modal 11th. boat ranks 130th, because each run counts against its plan's start limits. | 15th cheapest of 336 $140 a month | 13th cheapest of 336 $119 a month | 189th of 336, 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 Sandboxes that your users come back to: boat ranks 2nd, Upstash 3rd and Modal 15th. Upstash wakes a paused box on a request. On Modal your code saves and restores each user's filesystem snapshot. | 18th cheapest of 349 $671 a month | 4th cheapest of 349 $222 a month | 3rd cheapest of 349 $209 a month |
| RL rollouts 1,000 at once · 2 vCPU / 4 GB · 1,000,000 sessions of 3 min a month RL rollouts launch many short sandboxes at once. Modal ranks 6th, boat 29th and Upstash Box 206th; boat and Upstash are compromises here. | 6th cheapest of 349 $11,982 a month | 343rd of 349, with limits $8,000 a month | 39th of 349, with limits $2,000 a month |
| Agent evals 250 at once · 4 vCPU / 8 GB · 15,000 sessions of 20 min a month | 62nd cheapest of 359 $889 a month | 67th cheapest of 359 $1,011 a month | 10th cheapest of 359 $200 a month |
| CI runners 40 at once · 8 vCPU / 16 GB · 6,000 sessions of 12 min a month CI jobs that build Docker images: boat ranks 2nd, Modal 82nd and Upstash 90th. boat runs Docker in every sandbox, Modal with runtime="vm", and Upstash does not document it. | 110th cheapest of 374 $1,112 a month | 122nd cheapest of 374 $1,480 a month | 2nd cheapest of 374 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB Upstash ranks 5th with a flat-price keep-alive box, boat 13th, and Modal 84th as a compromise, because a Modal sandbox ends within 24 hours. | 129th of 371, with limits $9.07 a month | 8th cheapest of 371 $19 a month | 19th cheapest of 371 $26 a month |
| Computer-use agents 50 at once · 4 vCPU / 8 GB · 20,000 sessions of 15 min a month Modal ranks 6th, boat 8th and Upstash 76th. boat has a desktop built in. Modal's official example installs Xvfb, Chromium and noVNC in a VM sandbox. Upstash has a managed headless Chromium and no desktop. | 45th cheapest of 367 $153 a month | 123rd of 367, with limits $76 a month | 7th cheapest of 367 $200 a month |
| Browser-use agents 50 at once · 2 vCPU / 4 GB · 30,000 sessions of 10 min a month | 57th of 356, with limits $121 a month | 48th of 356, with limits $38 a month | 7th cheapest of 356 $200 a month |
| Scraping & bots (startup scale) 50 at once · 1 vCPU / 2 GB · 100,000 sessions of 3 min a month | 98th cheapest of 343 $252 a month | 41st cheapest of 343 $400 a month | 6th cheapest of 343 $90 a month |
| Mass scraping & bots (undetectable) 1,000 at once · 1 vCPU / 2 GB · 3,000,000 sessions of 2 min a month | 171st of 327, with limits $5,957 a month | 201st of 327, with limits $8,000 a month | 73rd of 327, with limits $1,800 a month |
| Parallel red teaming 200 at once · 2 vCPU / 4 GB · 5,000 sessions of 30 min a month | 289th of 351, with limits $464 a month | 255th of 351, with limits $259 a month | 3rd cheapest of 351 $200 a month |
| GPU eval jobs 10 at once · 8 vCPU / 32 GB · 500 sessions of 30 min a month Modal ranks 16th, with GPUs from T4 to B300. Neither Upstash Box nor boat has a GPU. | 18th cheapest of 102 $1,116 a month | Not ranked: misses a requirement | Not ranked: misses a requirement |
| macOS agents 5 at once · 4 vCPU / 8 GB · 1,000 sessions of 20 min a month | Not ranked: misses a requirement | Not ranked: misses a requirement | Not ranked: misses a requirement |
Key features
| Feature | Modal | Upstash Box | boat |
|---|---|---|---|
| Isolation | gVisor by default; VM runtime with its own kernel (runtime="vm") | Docker container per box on AWS | Full VM with its own kernel on hosts dedicated to boat; vCPUs are shared |
| Docker inside | Yes with the VM runtime; not documented on the default gVisor runtime | Not documented; the box is itself a Docker container | Yes, Docker and systemd in every sandbox |
| Nested virtualization | VM runtime only, Enterprise plan, as an experimental option | Not documented; boxes are containers | Yes, /dev/kvm for Android emulators and VMs |
| Stop and resume | No paused state; your code takes a filesystem snapshot (kept 30 days by default, or indefinitely) and starts a new sandbox from it | Pauses itself when idle and keeps its files; storage bills while paused | Disk saved every minute and at each stop; no compute charge while stopped; disk kept at least 14 days, longer with a saved snapshot |
| Longest session | 24 hours (default 5 minutes); a filesystem snapshot carries the disk to a new sandbox | No documented cap; keep-alive boxes stay on all month | No cap once auto-stop is off on a paid plan (default 1 h, timer up to 30 days) |
| Wake on request | No | Yes; a request to a public URL resumes a paused box | No; a stopped sandbox is started by API |
| Fork and snapshots | A filesystem snapshot boots many sandboxes; memory snapshots in alpha (kept 7 days) | Snapshot a running box and start many new boxes from it | Fork a running sandbox without stopping it, from its last 1-minute snapshot |
| Disk | 512 GiB by default | 10, 20 or 30 GB by size | 51 GB at 4 vCPU; up to 251 GB on bigger sizes |
| Custom images | Build images in code or pull any linux/amd64 registry image | Custom images listed as coming soon; reuse setups through snapshots | Start from boat's image or a saved snapshot; you cannot boot your own Docker image |
| GPUs | T4 to B300; 10 at once on Starter, 50 on Team | None | None |
| Desktop, browser, computer use | No built-in desktop; the official example installs Xvfb, Chromium, x11vnc and noVNC in a VM sandbox | Managed headless Chromium with AI page actions and CDP; no desktop or VNC | 1920x1080, 60 fps desktop, Chrome and computer tools built in |
| Coding agents | You install a harness in the image (official OpenCode example) | Claude Code, Codex, OpenCode and Cursor built in, with run and stream APIs, webhooks and scheduled prompts | Seven harnesses preinstalled, with a prompt, steer and events API |
| Keeping API keys out of the sandbox | Secrets are environment variables, or you run your own proxy in a sidecar (beta) | Host proxy injects headers into outbound HTTPS on port 443 for any host, set at creation | Optional paid AI credentials proxy ($1 a day + $1 per 1,000 requests) for supported model providers |
| Outbound network rules | Block all, IP-range allowlist, domain allowlist (beta) | Allow or deny by domain and CIDR, changeable on a running box | No provider-side allowlist |
| Start limits | No start-rate cap published; 100 containers at once on Starter, 5,000 on Team | No start limit published; 1,000 boxes at once on Pay as You Go | $20 plan: 12 a minute, 200 a day included; $2,000 plan: 90 a minute, 2,400 a day; paid extra starts up to 3x per minute, then refused |
| Regions | US, EU and Asia-Pacific; pinning costs 1.15x to 1.75x | AWS us-east-1; custom regions on Enterprise | Germany, Finland and France; no US or Asia region, no region choice |
Every feature Battleships tracks (43)
| Feature | Modal | Upstash Box | boat |
|---|---|---|---|
| Full VM (own kernel) | No documented way | No documented way | Yes |
| Docker inside | No documented way | Not published | Yes |
| Nested virtualization | No documented way | No documented way | Yes |
| Root | Yes | Not published | Yes |
| Systemd | No documented way | Not published | Yes |
| Desktop GUI | With your own setup | No documented way | Yes |
| Browser | With your own setup | Yes | Yes |
| Computer-use API | With your own setup | No documented way | Yes |
| SSH | With your own setup | Yes | Yes |
| Persistent disk | With your own setup | Yes | Yes |
| Pause/resume | No documented way | Yes | Yes |
| Automatic snapshots | No documented way | No documented way | Yes |
| Fork/clone | Yes | Yes | Yes |
| Live fork (no pause) | Not published | Yes | Yes |
| Memory fork | Yes | No documented way | No documented way |
| Custom image (Docker or snapshot) | Yes | Yes | Yes |
| ≥24 h sessions | Yes | Yes | Yes |
| Idle auto-stop | Yes | Yes | No documented way |
| Wake on request | No documented way | Yes | No documented way |
| HTTPS preview URLs | Yes | Yes | Yes |
| Raw TCP inbound | Yes | No documented way | Yes |
| Public IPv4 | No documented way | No documented way | No documented way |
| Egress allowlist | Yes | Yes | No documented way |
| Secret proxy | No documented way | Yes | Yes |
| Secret proxy for any API | No documented way | Yes | Yes |
| Preinstalled agents | With your own setup | Yes | Yes |
| Agent harness API | No documented way | Yes | Yes |
| MCP server | No documented way | Yes | No documented way |
| GPU | Yes | No documented way | No documented way |
| Windows | No documented way | No documented way | No documented way |
| Arm64 | No documented way | Yes | No documented way |
| Self-hosting | No documented way | No documented way | No documented way |
| BYOC | No documented way | No documented way | No documented way |
| Open source | No documented way | No documented way | No documented way |
| Scoped API keys | Not published | Not published | Yes |
| Spend limits | Not published | Not published | Yes |
| Usage per sandbox (API) | Not published | Yes | Yes |
| Audit logs | Yes | Not published | No documented way |
| SSO | Yes | No documented way | No documented way |
| SOC 2 | Yes | Yes | No documented way |
| HIPAA | Yes | No documented way | No documented way |
| EU data residency | Yes | No documented way | No documented way |
| Zero data retention | Not published | Not published | Yes |
Performance
StarSling's run of 3 October 2026 (run 37073680030) ran the V8 Web Tooling Benchmark at 4 vCPU / 8 GB. Modal's default gVisor runtime did 12.60 runs/s and its VM runtime 15.48. boat did 28.06, second of 15. The boat sandbox ran on an AMD Ryzen 9 9950X host; boat also places sandboxes on other hosts. Upstash Box is not in that run, so we cannot say how it compares. Providers ran on different hardware in different regions.
V8 Web Tooling Benchmark at a 4 vCPU / 8 GB target, runs per second, higher is better. Source: StarSling's run of 3 October 2026 (run 37073680030), median of 5 sandboxes. Providers ran on different hardware in different regions. StarSling gives a shared rank when the run cannot separate two providers. Modal measured on: Sandbox (gVisor), request = size, any region. Every boat sandbox in this run ran on an AMD Ryzen 9 9950X host. boat also places some sandboxes on AMD Ryzen 7 9700X hosts, which this run did not measure. Upstash Box is not in the public benchmarks we use, so we cannot say which is faster.
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. Start limits, per-start fees and plan minimums are in the price, boat's included.
| Scenario | Modal, a month | Upstash Box, a month | boat, 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 | $0 · Starter | $21 · Pay as You Go | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $109 · Starter · with limits | $21 · Pay as You Go | $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 | $1,480 · Starter · with limits | $900 · Pay as You Go · with limits | $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 | $579 · Team | $306 · Pay as You Go | $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 | $1,911 · Team · with limits | $1,822 · Pay as You Go | $8,800 · $2000/mo · with limits |
Modal: Modal bills by the second for whichever is higher, the CPU and memory you request or what you use. An idle sandbox pays until it is terminated. Modal prices CPU per physical core, and one core equals 2 vCPU, so a 4 vCPU / 8 GiB sandbox is $0.476 an hour at a full request. Starter is free with $30 of monthly credit, Team is $250 a month with $100 of credit, and pinning a region multiplies the price by 1.15x to 1.75x.
Upstash Box: A standard box is billed for the CPU it actually uses, at $0.10, $0.20 or $0.40 per active CPU hour for Small, Medium and Large, plus $0.10 per GB-month of storage. RAM is not billed. A keep-alive box costs a flat $8, $16 or $32 a month whether it is busy or not.
Limits in those scenarios
A Modal sandbox lives at most 24 hours. Your code keeps its disk with a filesystem snapshot or a Volume and starts a new sandbox from it. Docker needs Modal's VM runtime, and Upstash does not document Docker inside a box, so the 50-agent fleet fits only Modal's VM runtime ($1,480 a month) and boat ($317). Upstash does not publish its egress price, and its keep-alive pool is billed whether used or not. boat's limits: plans start at $20 a month, there is no GPU, and there is no US or Asia region. A running boat sandbox bills at full rate, busy or idle. A stopped sandbox has no compute charge, but extra saved snapshots and the optional AI credentials proxy still charge. For 1,000 short jobs at once, Upstash fits at $1,822 a month. boat charges for starts above 90 a minute and refuses starts above 270 a minute, so that scenario does not fit boat.
| Scenario | Modal | Upstash Box | boat |
|---|---|---|---|
| One coding agent, workdays | None for this scenario |
| None for this scenario |
| One agent that never stops |
|
| None for this scenario |
| 50 agents in parallel |
|
| None for this scenario |
| Sandboxes for your users | None for this scenario |
| None for this scenario |
| 1,000 short jobs at once |
|
|
|
Which one should you use?
Use Modal when
your sandboxes need GPUs, must run outside the US, need a big disk or your own image, or are huge bursts of short jobs.
- GPUs on the same platform. Useful if your agent trains or runs models: Modal rents GPUs from T4 to B300 next to its sandboxes.
- Regions around the world. Modal runs in the US, EU and Asia-Pacific. Upstash Box runs in AWS us-east-1 on self-serve plans.
- Huge fleets, big disks and any image. Modal runs up to 5,000 containers on Team, gives each sandbox 512 GiB of disk and boots any linux/amd64 registry image.
Use Upstash Box when
your agents mostly wait, need to wake on an incoming request, keep their files with no code, and can run in the US without Docker.
- Idle costs no CPU, and a request wakes the box. Good for agents that wait a lot: an idle box costs no CPU, RAM is not billed, and a request to its URL wakes it.
- Files kept with no code. A box keeps its files across pauses with no documented session cap. On Modal your code takes a filesystem snapshot before the 24-hour limit.
- Coding agents built in. Claude Code, Codex, OpenCode and Cursor come ready to run, with webhooks and scheduled prompts. On Modal you install them in the image.
Where boat fits
boat fits coding and computer-use agents that work for hours on one machine. Each sandbox is a full VM with Docker, a desktop and a disk that boat saves with no code. Modal fits better for GPUs, regions and huge bursts. Upstash fits better for agents that mostly wait.
Docker and /dev/kvm in every sandbox
boat runs Docker and /dev/kvm in every sandbox with no setting. Modal runs Docker with runtime="vm", and nested virtualization is an experimental Enterprise option. Upstash does not document Docker inside a box, so the 50-agent scenario does not fit there.
A desktop the agent can see
boat streams a 1920x1080, 60 fps desktop with Chrome and computer tools. On Modal you build that image yourself; its official example installs Xvfb, Chromium and noVNC. Upstash gives a managed headless Chromium and no desktop.
A disk kept with no code
boat saves the disk every minute and at each stop, and keeps a stopped disk at least 14 days. On Modal your code takes a filesystem snapshot. Upstash keeps files with no code, but its disk is 10 to 30 GB. boat has 51 GB at 4 vCPU. If a boat machine fails, you lose what was written after the last 1-minute snapshot.
Use boat when your agents need a full machine for hours: Docker, emulators, a desktop and a disk saved with no code. Turn auto-stop off on a paid plan for long work, and check your plan's start limits. Compare it one to one: boat vs Modal and boat vs Upstash Box.
boat's own limits
- Plans start at $20 a month. The plan price comes back as VM time for that month. Unused time does not carry over. Source
- Starts (create, fork, resume): the $20 plan includes 12 a minute, 60 an hour and 200 a day. The $2,000 plan includes 90 a minute, 600 an hour and 2,400 a day. Extra starts cost $0.03 to $0.07. A start fails with 429 above 3 times the included starts per minute, or 6 times per hour or per day. Source
- A stopped sandbox has no compute charge. boat keeps its disk for at least 14 days. A saved snapshot keeps it longer; extra saved snapshots and the AI credentials proxy are paid. Source
- Auto-stop defaults to 1 hour. You can turn it off on a paid plan. The trial allows 2 hours at most. Source
- boat saves the disk every minute and at each stop. Memory, running programs and /tmp are not saved: a resume is like a reboot. A machine failure loses writes after the last good snapshot. Source
- Machines run in Germany, Finland and France only. Snapshot copies can be stored outside the EU. Source
- Public IPv6 and outbound IPv4. No dedicated IPv4. The IP changes at each resume. Source
- Hosts are AMD Ryzen 7 9700X and Ryzen 9 9950X. vCPUs are shared. No GPU. You cannot boot your own Docker image. Source
Questions people ask
Is Modal cheaper than Upstash Box?
It depends on the workload. One workday agent costs $0 on Modal inside its $30 monthly credit, against $21 on Upstash. Sandboxes for your users cost $306 on Upstash and $579 on Modal. One agent that never stops costs $21 on Upstash and $109 on Modal, but Modal ends each sandbox at 24 hours. 1,000 short jobs at once cost $1,822 on Upstash and $1,911 on Modal. On boat, the first four cost $20, $200, $26 and, for 50 agents, $317.
Which is faster, Modal or Upstash Box?
Upstash Box is not in the public CPU benchmark we use, so we cannot say. In StarSling's run of 3 October 2026, Modal did 12.60 runs/s on gVisor and 15.48 on its VM runtime at 4 vCPU / 8 GB. boat did 28.06 on a Ryzen 9 9950X host. Providers ran on different hardware.
Can Modal or Upstash Box run Docker?
Modal runs Docker on its VM runtime, which you choose with runtime="vm". Upstash Box does not document it: each box is itself a Docker container. Every boat sandbox runs Docker and systemd with no extra setting.
Can a Modal sandbox keep its files?
Yes, if your code does it. A filesystem snapshot saves the root disk and boots new sandboxes, and it is kept 30 days by default or indefinitely. A Volume keeps one folder. Upstash keeps a box's files across pauses with no code. boat saves the disk every minute and at each stop with no code.
How long can a sandbox run on Modal and Upstash Box?
A Modal sandbox defaults to 5 minutes and lasts at most 24 hours. An Upstash box has no documented cap: it pauses when idle and resumes later, and a keep-alive box stays on all month. boat has no session cap once auto-stop is off on a paid plan (default 1 hour).
Do Modal or Upstash Box have GPUs?
Modal does, from T4 up to B300, on its default gVisor runtime. Upstash Box has none, and boat sandboxes are CPU machines too, so for GPU work Modal is the fit.
Where do Modal and Upstash Box run?
Modal runs in the US, the EU and Asia-Pacific, and you can pin a sandbox to a region for a higher price. Upstash Box runs in AWS us-east-1 on self-serve plans, with custom regions on Enterprise. boat runs in Germany, Finland and France, and you cannot pick a region.
Try boat
Install the CLI, then run boat new.
7-day free trial