Steel vs Scrapeless
Both rent you a cloud Chromium that your agent drives over CDP. Steel is an open-source browser API with saved logins, human takeover and hour-long sessions on Scale, while Scrapeless is a scraping company whose browser comes with fingerprints, residential proxies and CAPTCHA solving built in.
The companies and products
Steel
Steel is an API that runs fleets of cloud browsers for AI agents: you create a session, connect Puppeteer, Playwright or Selenium, and release it when done.
Each Steel session is a headful Chromium in a VM in Steel's cloud, driven over CDP or through a computer-use input API, with a live viewer you can embed. Around it Steel sells managed proxies, CAPTCHA solving, saved browser profiles, a credentials vault that signs in for the agent, a files API and readable traces. It is a different kind of product from bochka: a managed browser, not a general sandbox, so this page compares where the two overlap, which is agents that use a browser.
Scrapeless
Scrapeless is a web scraping company that sells Agent Browser, a managed cloud browser, along with AI scrapers, a web unlocker and proxies.
Scrapeless Agent Browser is a different kind of product from bochka: you get a Chromium browser in their cloud that you reach over a CDP WebSocket with Puppeteer or Playwright, an MCP server or their agent-browser CLI. Each session gets a random fingerprint, CAPTCHA solving and a residential proxy, and you pay per browser-hour plus proxy traffic. It is built for teams that scrape or automate many websites, not a machine to run their own code on.
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. “25th cheapest of 367” means 24 of the other 366 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 | Steel: price rank | Scrapeless: price rank | bochka: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Neither gives you a shell, Docker or a place to build and test, so both are a compromise here, while bochka fits fully. | 124th of 367, with limits $1,115 a month | 96th of 367, with limits $868 a month | 3rd cheapest of 367 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB An assistant that works all month hits Steel's 1 hour cap on Scale and Scrapeless's minutes-long sessions, so both keep restarting from a saved profile. | 238th of 366, with limits $3,070 a month | 224th of 366, with limits $2,628 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 | 264th of 345, with limits $333 a month | 246th of 345, with limits $249 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 300 sandboxes at once is more than Steel's Scale limit of 100, and Scrapeless sells no machine your users can run code on. | 244th of 361, with limits $990 a month | 354th of 361, with limits $741 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 | 257th of 360, with limits $4,150 a month | 249th of 360, with limits $3,580 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 | 348th of 364, with limits $400 a month | 347th of 364, with limits $399 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 CI needs a shell, Docker and a known CPU size, which neither browser service offers. | 239th of 377, with limits $120 a month | 229th of 377, with limits $97 a month | 4th cheapest of 377 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB | 241st of 368, with limits $73 a month | 234th of 368, with limits $59 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 Both drive only a browser; bochka streams a 60fps Linux desktop where the agent can use any app as well as Chromium. | 148th of 369, with limits $367 a month | 122nd of 369, with limits $274 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 Both fit browser-use agents; Scrapeless ranks 14th and Steel ranks 25th, and bochka ranks 7th because the agent also gets a shell and a disk next to Chromium. | 25th cheapest of 367 $470 a month | 14th cheapest of 367 $274 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 At startup scale both fit, with Scrapeless ahead of Steel; bochka ranks 6th by running the scraper, its queue and its database on one machine. | 56th cheapest of 346 $540 a month | 33rd cheapest of 346 $370 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 For mass scraping of sites with strong bot checks, Scrapeless ranks 6th and Steel ranks 9th, well ahead of bochka, which has datacenter IPs and no CAPTCHA solving. | 9th of 338, with limits $7,950 a month | 6th of 338, with limits $7,020 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 | 220th of 360, with limits $242 a month | 200th of 360, with limits $199 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 | Steel | Scrapeless | bochka |
|---|---|---|---|
| Kind of product | A managed Chromium browser per session, driven over CDP or a computer-use input API | A managed Chromium browser per session over a CDP WebSocket, sold by a scraping company | A full Linux VM per sandbox with its own kernel, and a desktop and Chromium inside |
| Machine size | Not published; billed per browser hour | Not published; billed per browser hour | 4 vCPU / 8 GB by default, up to 16 vCPU / 32 GB by plan |
| Shell, Docker and nested virtualization | No shell; CDP and a REST API only | No shell on the machine running the browser | SSH, root, systemd, Docker and /dev/kvm on every machine |
| Session length | 15 minutes on Launch, 1 hour on Scale, up to 24 hours on Enterprise | 3 minutes by default, 15 minutes recommended at most, longer allowed | No session cap on paid plans |
| Concurrency | 10 browsers on Launch, 100 on Scale, 1,000+ on Enterprise | 50, 100, 200 or 400 sessions by plan, custom above that | Set by plan; bursts above 90 starts a minute are queued |
| State between sessions | Profiles keep cookies and logins; a files API keeps session files | Browser profiles keep cookies, storage, cache and logins, saved when the session closes | The whole disk is saved every minute and on stop; a stopped sandbox costs nothing |
| Fork | No fork; a new session can start from a saved profile | No fork; sessions can share a profile, and only the last to close is saved | Fork a running sandbox into a copy |
| Stealth and CAPTCHAs | CAPTCHA solving on paid plans, billed per solve; a stealth browser on Enterprise | Random fingerprint per session and CAPTCHA solving included at no extra charge | Chromium on datacenter IPs; no CAPTCHA solving |
| Proxies | Managed residential proxies billed per GB, dedicated IPs on Scale, your own proxy on every plan | Residential IPs in 195 countries down to the city; your own proxy on paid plans | None managed; you can run your own proxy client inside the VM |
| Live view and human takeover | Embeddable live view where a person can take over, mobile mode and traces | Live URL to watch and control the browser, plus optional session recording | A full Linux desktop streamed at 60fps, embeddable, with computer use |
| Open source and self-hosting | steel-browser is Apache-2.0 open source; Steel Outpost runs inside your network | Managed cloud only | Managed cloud only |
| Ready-made scrapers | Browser tools for scrape, screenshot and PDF, billed per call | AI Scraper tasks for ChatGPT, Perplexity, Gemini and more, plus crawl and search over MCP | None; you write or run your own |
| Reaching the machine | No inbound ports; a CDP websocket and a live view URL | No way to serve your own ports | HTTPS URLs for up to 50 ports, token-protected; raw TCP |
| Agents and secrets | CLI, skills and an MCP server for an agent that runs elsewhere; a credentials vault for web logins | CLI skill and an MCP server for an agent that runs elsewhere; a credential store and 1Password link | Coding agents preinstalled with a prompt, steer and events API; a credentials proxy for any HTTPS API |
Every feature Battleships tracks (43)
| Feature | Steel | Scrapeless | bochka |
|---|---|---|---|
| Full VM (own kernel) | Yes | Not published | Yes |
| Docker inside | No | No | Yes |
| Nested virtualization | No | No | Yes |
| Root | No | Not published | Yes |
| Systemd | No | Not published | Yes |
| Desktop GUI | No | No | Yes |
| Browser | Yes | Yes | Yes |
| Computer-use API | Yes | Yes | Yes |
| SSH | No | No | Yes |
| Persistent disk | No | No | Yes |
| Pause/resume | No | No | Yes |
| Automatic snapshots | Yes | No | Yes |
| Fork/clone | No | No | Yes |
| Live fork (no pause) | No | No | Yes |
| Memory fork | No | No | No |
| Custom image (Docker or snapshot) | No | Not published | Yes |
| ≥24 h sessions | Yes | No | Yes |
| Idle auto-stop | Yes | Yes | No |
| Wake on request | No | No | No |
| HTTPS preview URLs | No | No | Yes |
| Raw TCP inbound | No | Not published | Yes |
| Public IPv4 | No | No | No |
| Egress allowlist | No | No | No |
| Secret proxy | No | No | Yes |
| Secret proxy for any API | No | No | Yes |
| Preinstalled agents | No | No | Yes |
| Agent harness API | No | No | Yes |
| MCP server | Yes | Yes | No |
| GPU | No | No | No |
| Windows | No | No | No |
| Arm64 | No | No | No |
| Self-hosting | Yes | No | No |
| BYOC | Yes | No | No |
| Open source | Yes | No | No |
| Scoped API keys | Not published | Not published | Yes |
| Spend limits | Not published | Not published | Yes |
| Usage per sandbox (API) | No | No | Yes |
| Audit logs | Not published | Not published | No |
| SSO | Yes | Yes | No |
| SOC 2 | No | Yes | No |
| HIPAA | Yes | Yes | No |
| EU data residency | No | No | Yes |
| Zero data retention | Not published | Not published | Yes |
Performance
There is no comparable measurement for either. Steel and Scrapeless sell browser hours, do not publish the CPU or memory behind each session, and are not in StarSling's CPU benchmark. bochka scored 26.69 runs per second on its Node.js test at 4 vCPU / 8 GB.
Node.js build-and-test loop at 4 vCPU / 8 GB, higher is better. Measured with the open hpc-sandbox-benchmarks suite. Steel and Scrapeless 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 | Steel, a month | Scrapeless, 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 | $250 · Scale · with limits | $399 · Business · with limits | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $250 · Scale · with limits | $399 · Business · with limits | $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 | $784 · Scale · with limits | $570 · Business · 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 | $510 · Scale · with limits | $399 · Business · 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 | $950 · Scale · with limits | $716 · Business · with limits | $2,000 · $2000/mo · with limits |
Steel: Steel bills browser hours by the minute, rounded up: $0.10 an hour on the free Launch plan and $0.08 on Scale, which costs $250 a month and includes $100 of usage. Proxy bandwidth, CAPTCHA solves and browser tools are billed on top.
Scrapeless: Scrapeless plans of $49, $199 or $399 a month are prepaid into your balance, and unused funds do not roll over. Browser time runs from $0.09 an hour down to $0.072, CAPTCHA solving is included, and residential proxy traffic is billed per GB from $1.80.
Scrapeless: Plans start at $20 a month, and the whole fee is usage credit spent per second on running sandboxes. A stopped sandbox costs nothing and keeps its disk.
Limits in those scenarios
None of the five scenarios fits Steel or Scrapeless, because each asks for a 4 vCPU / 8 GB machine with a disk and they sell a browser session with no shell and no published size. Steel's Scale plan stops sessions after 1 hour and allows 100 browsers at once, so 300 user sandboxes or 1,000 short jobs need Enterprise. Scrapeless sessions are meant to last minutes and its biggest self-serve plan stops at 400 at once. On bochka, only the 1,000-job burst does not fit: bursts above 90 starts a minute are queued.
| Scenario | Steel | Scrapeless | 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 Steel when
you want an open-source browser you can host yourself, sessions up to an hour on Scale, and saved logins, human takeover and traces as ready parts.
- Open source, and can run in your network. For teams that must host browsers themselves: steel-browser is Apache-2.0, and Steel Outpost runs inside your own network.
- Longer sessions on a plan you can buy. For agent runs past 15 minutes: Scale sessions last up to 1 hour, and Enterprise up to 24 hours.
- Ready browser parts for agents. For agents that sign in and need a person nearby: saved profiles, a credentials vault, human takeover and readable traces.
Use Scrapeless when
you run many short sessions on sites with strong bot checks and want fingerprints, residential proxies and CAPTCHA solving included.
- Stealth included. For sites with strong bot checks: every session gets a random fingerprint, and CAPTCHA solving costs nothing extra.
- Residential proxies everywhere. For scraping by location: over 70 million residential IPs in 195 countries, which you can pin to a state or city.
- Data, not only browsers. For teams that want results: ready-made AI scrapers, crawls and search tools sit behind the same API key.
Where bochka fits
Steel and Scrapeless are both browser services, so bochka only overlaps where your agent uses a browser. If what you are really running is agents that browse and then work, writing files, running code, building in Docker or serving an app, look at bochka: each agent gets a whole Linux VM with Chromium and a desktop inside.
A whole machine around the browser
Neither Steel nor Scrapeless gives you a shell, so installs, builds, Docker and tests run on a second machine you host. You wire the two together, move every file between them and keep both alive.
No session cap, and a disk that stays
Steel stops a Scale session after 1 hour, and Scrapeless sessions default to 3 minutes. Both carry only a browser profile forward, so files on disk and running work start over each time; bochka saves the whole disk every minute.
Agents inside, driven from outside
Both offer an MCP server so an agent on your own laptop or server can use their browser, which means you still host that agent and its model keys. On bochka the agent runs inside, behind a credentials proxy, and you drive it through one API.
Use bochka when your agent does more than browse: it needs a shell, files, Docker, a desktop or a server next to Chromium, and it has to keep working for hours. Compare it one to one: bochka vs Steel and bochka vs Scrapeless.
Questions people ask
Is Steel cheaper than Scrapeless?
It depends on the job. For one coding agent on workdays Steel comes to $250 a month against $399 on Scrapeless, but for 50 agents in parallel Steel comes to $784 against $570, and for 1,000 short jobs at once $950 against $716. Neither fits these scenarios as a general machine, and bochka comes to $20, $317 and $2,000 for the same three.
Which is better at getting past bot checks?
Scrapeless includes a random fingerprint per session and CAPTCHA solving at no extra charge, with residential IPs in 195 countries. Steel has residential proxies and CAPTCHA solving billed per use, and keeps its stealth browser for Enterprise. bochka uses datacenter IPs and solves no CAPTCHAs.
How long can a session run on Steel and Scrapeless?
Steel allows 15 minutes on Launch, 1 hour on Scale and 24 hours on Enterprise. Scrapeless defaults to 3 minutes and recommends 15 minutes at most, though longer values can be set. bochka has no session cap on paid plans.
Can Steel or Scrapeless run Docker or a coding agent?
No. Both give you a cloud Chromium with no shell, so Docker and an agent's code must run on a machine you host, which then drives the browser through their MCP server or API. On bochka each sandbox is a full VM with Docker and coding agents preinstalled.
Which is faster?
Neither publishes the CPU behind a session, and neither is in the CPU benchmark, so there is no fair speed number. Steel says a session starts in under a second, and Scrapeless says under 2 seconds.
Can I self-host Steel or Scrapeless?
Steel, yes: steel-browser is open source under Apache-2.0 and runs in Docker, and Steel Outpost runs inside your own network. Scrapeless is a managed cloud only.
Try bochka
Install the CLI, then run bochka new.
Product preview