Runloop vs Claude Managed Agents
Runloop rents you devboxes, micro-VMs where you run Claude Code, Codex or your own agent. Claude Managed Agents is Anthropic's hosted agent: you send Claude a task, and it works in a container that Anthropic runs, or on a worker that you host.
The companies and products
Runloop
Runloop is a US platform of sandboxes, called devboxes, for building, running and testing AI coding agents.
You get Linux devboxes built from Dockerfile-style blueprints, disk snapshots, HTTPS tunnels, agent gateways that hide API keys, browser and computer use, and Axons, event streams that drive agents such as Claude Code. Each devbox runs in a micro-VM on Runloop's own bare-metal hypervisor, with a container layer inside. It is built for teams that ship coding agents, plus evaluation and RL fine-tuning work on the Enterprise plan.
- Architectures
- x86_64 and arm64 base blueprints
- Own-cloud deploy
- Deploy to VPC on AWS, GCP and Azure (Enterprise)
- Regions
- Not documented or selectable
- Company
- Runloop AI, Inc.
Claude Managed Agents
Claude Managed Agents is Anthropic's ready-made agent harness: Claude runs as an autonomous agent, and you talk to it through an events API.
This is an agent platform, not a sandbox you rent. You define an agent (model, prompt, tools, MCP servers, skills) and an environment, then start sessions, send messages as events and stream back what Claude does. The harness handles tool calls, prompt caching and compaction. Each session gets its own Ubuntu container in Anthropic's cloud, or runs its tools on a worker that you host. It is built for teams that want Claude to do long tasks without building their own agent loop or sandbox.
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. “24th cheapest of 365” means 23 of the other 364 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 | Runloop: price rank | Claude Managed Agents: price rank | boat: price rank |
|---|---|---|---|
| Software factory 100 at once · 4 vCPU / 8 GB · 2,200 sessions of 8 h a month Runloop gives a coding agent a devbox with Docker on its dnd blueprints. A Managed Agents cloud sandbox has up to 8 GB of memory and Docker in limited availability; a self-hosted worker can be a bigger host that you run. boat adds /dev/kvm and saves the disk every minute. | 24th cheapest of 365 $8,152 a month | 139th of 365, with limits $964 a month | 5th cheapest of 365 $634 a month |
| Always-on AI employees 50 machines 24/7 · 2 vCPU / 4 GB Both can sleep an idle agent and wake it: Runloop on the Pro plan through an HTTP request, Managed Agents on the next message. Managed Agents keeps cloud sandbox files 30 days after creation; history and memory stores stay until you delete them. | 38th cheapest of 361 $10,194 a month | 259th of 361, with limits $2,511 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 Managed Agents bills runtime only while the agent works, and Runloop publishes no start-rate limit on paid plans. boat's start limits make large bursts cost more. | 52nd cheapest of 336 $525 a month | 167th of 336, with limits $756 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 Your users can open a Runloop devbox through its HTTPS tunnel. A Managed Agents cloud sandbox has no documented inbound ports; with a self-hosted worker you open ports on your own host. boat gives each sandbox HTTPS URLs on up to 50 ports. | 37th cheapest of 349 $3,301 a month | 227th of 349, with limits $823 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 Runloop ranks 8th for RL rollouts. It publishes no start-rate limit on paid plans and sells RL fine-tuning on Enterprise. | 8th cheapest of 349 $15,922 a month | 247th of 349, with limits $3,978 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 | 91st cheapest of 359 $2,562 a month | 247th of 359, with limits $320 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 Runloop runs Docker-in-Docker for builds, and the Managed Agents cloud sandbox lists Docker as limited availability. In StarSling's run of 3 October 2026, boat did 28.06 runs a second and Runloop 8.93. | 125th cheapest of 374 $1,554 a month | 228th of 374, with limits $96 a month | 2nd cheapest of 374 $86 a month |
| 24/7 dev boxes 1 machine 24/7 · 4 vCPU / 8 GB A dev box needs SSH and a lasting disk. Runloop has SSH through a TLS proxy. The Managed Agents cloud sandbox has no documented SSH. boat has SSH and keeps a stopped disk for at least 14 days, or longer as a saved snapshot. | 118th cheapest of 371 $410 a month | 238th of 371, with limits $48 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 Runloop ships browser and computer use. Managed Agents ships headless Chromium with Playwright; you add a desktop through custom tools, MCP or a self-hosted worker. boat streams a full Linux desktop at 60 fps. | 41st cheapest of 367 $2,306 a month | 166th of 367, with limits $288 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 | 38th cheapest of 356 $1,153 a month | 181st of 356, with limits $288 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 | 73rd cheapest of 343 $781 a month | 194th of 343, with limits $390 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 | 267th of 327, with limits $15,611 a month | 153rd of 327, with limits $4,798 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 | 16th cheapest of 351 $770 a month | 219th of 351, with limits $160 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 | Not ranked: misses a requirement | 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 | Runloop | Claude Managed Agents | boat |
|---|---|---|---|
| What it is | A sandbox platform: devboxes for building, running and testing your coding agents | A hosted Claude agent: Anthropic's harness runs Claude, and tools run in a cloud sandbox or on a worker you host | A sandbox platform: each agent gets its own full Linux VM, with coding agents installed |
| Isolation | Micro-VM on a custom bare-metal hypervisor, plus a container layer | Cloud sandbox: isolated Linux container per session. Self-hosted worker: the isolation of the host you run | Full VM with its own kernel, on physical hosts dedicated to boat (AMD Ryzen 7 9700X and Ryzen 9 9950X); vCPUs are shared |
| Machine size | Sizes such as 4 vCPU / 8 GiB and 2 vCPU / 4 GiB; disk up to 64 GiB | Cloud sandbox: up to 8 GB memory and 10 GB disk, vCPU not published, not configurable. Self-hosted worker: whatever your host has | 2 to 16 vCPU and 4 to 32 GB; 16 vCPU needs the $100 plan or higher; disk fixed per size, 51 to 251 GB |
| Docker, systemd and /dev/kvm | Docker as root on the dnd blueprints; nested virtualization not documented | Cloud sandbox: Docker listed as limited availability; root, systemd and /dev/kvm not documented. Self-hosted worker: whatever your host has | Root, Docker, systemd and /dev/kvm on every sandbox, for Android emulators and VMs |
| CPU architecture | x86_64 and arm64 | Cloud sandbox: x86_64 (Ubuntu 24.04) | x86_64 only |
| Keeping files | Disk snapshots on every plan; automatic idle suspend keeps the disk on Pro; shutdown deletes the disk | Cloud sandbox checkpointed when idle; its files kept 30 days after creation; history, memory stores and outputs kept until you delete them | Disk saved every minute and at each stop, on every plan; kept at least 14 days after a stop; saved snapshots have no scheduled expiry |
| Long runs | Keep-alive timer 1 hour by default, 48 hours max; an idle policy replaces it | Long sessions that sleep between messages; no session length cap published | Auto-stop 1 hour by default, up to 30 days, or off on a paid plan (then no session cap) |
| Idle sleep and wake | Suspend after idle, wake on an HTTP request to the tunnel (Pro) | Idle session resumes from its checkpoint on the next message; idle time is not billed | No idle timer; a stopped sandbox is resumed by API, like a reboot |
| Inbound access | One HTTPS tunnel per devbox covering its ports; SSH through a TLS proxy; no public IP documented | Cloud sandbox: no documented inbound ports or SSH; reached through the sessions and events API. Self-hosted worker: the ports you open on your host | Up to 50 HTTPS URLs, token-gated by default; SSH; public IPv6 with raw ports (IPv6 only on sandboxes without their own IPv4) |
| Browser and desktop | Browser and computer use for agents | Headless Chromium with Playwright; no built-in desktop or computer-use tool (add one through custom tools, MCP or a self-hosted worker) | Built-in Linux desktop and Chrome streamed at 1920x1080, 60 fps, with computer tools for agents |
| Agents | Claude Code, Codex, OpenCode, Gemini CLI and DeepAgents through public agent wrappers | Anthropic's harness with Claude models, with steering, memory stores, multiagent threads, session budgets and webhooks | Claude Code, Codex, pi, OpenCode, Prime Agent, Kimi Code, Mistral Vibe, driven from outside with a prompt, steer and events API |
| Keeping API keys out of the sandbox | Agent gateways for any API endpoint with bearer, header or basic auth | Vault secrets live in the sandbox as placeholders and are swapped in at egress, per host; not yet on self-hosted workers | Optional AI credentials proxy ($1 a day it is on plus $1 per 1,000 requests) for supported model providers and any HTTPS API; other secrets still enter the sandbox |
| Outbound rules | Hostname allowlist with wildcards, or block all | Unrestricted, or limited to listed hosts plus optional package registries and MCP servers | Open internet, no provider-side allowlist; you set up firewall rules inside the VM |
| Your own infrastructure | Deploy to VPC on AWS, GCP or Azure (Enterprise) | Self-hosted workers run tools on your own hosts while Anthropic keeps the orchestration | No; boat runs every sandbox on its own machines in Germany, Finland and France |
Every feature Battleships tracks (43)
| Feature | Runloop | Claude Managed Agents | boat |
|---|---|---|---|
| Full VM (own kernel) | Yes | No documented way | Yes |
| Docker inside | Yes | Not published | Yes |
| Nested virtualization | No documented way | No documented way | Yes |
| Root | Yes | Not published | Yes |
| Systemd | Not published | Not published | Yes |
| Desktop GUI | Yes | No documented way | Yes |
| Browser | Yes | Yes | Yes |
| Computer-use API | Yes | No documented way | Yes |
| SSH | Yes | No documented way | Yes |
| Persistent disk | Yes | Yes | Yes |
| Pause/resume | Yes | Yes | Yes |
| Automatic snapshots | Yes | Yes | Yes |
| Fork/clone | Yes | No documented way | Yes |
| Live fork (no pause) | Yes | No documented way | Yes |
| Memory fork | No documented way | 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 | Yes | Yes | No documented way |
| HTTPS preview URLs | Yes | No documented way | Yes |
| Raw TCP inbound | No documented way | No documented way | Yes |
| Public IPv4 | No documented way | No documented way | No documented way |
| Egress allowlist | Yes | Yes | No documented way |
| Secret proxy | Yes | Yes | Yes |
| Secret proxy for any API | Yes | Yes | Yes |
| Preinstalled agents | Yes | Yes | Yes |
| Agent harness API | Yes | Yes | Yes |
| MCP server | Yes | Yes | No documented way |
| GPU | No documented way | No documented way | No documented way |
| Windows | No documented way | No documented way | No documented way |
| Arm64 | Yes | No documented way | No documented way |
| Self-hosting | No documented way | Yes | No documented way |
| BYOC | Yes | Yes | No documented way |
| Open source | No documented way | No documented way | No documented way |
| Scoped API keys | Yes | Not published | Yes |
| Spend limits | Not published | Yes | Yes |
| Usage per sandbox (API) | Yes | Yes | Yes |
| Audit logs | Not published | Not published | No documented way |
| SSO | Yes | Yes | No documented way |
| SOC 2 | Yes | Yes | No documented way |
| HIPAA | Not published | No documented way | No documented way |
| EU data residency | No documented way | No documented way | No documented way |
| Zero data retention | Not published | No documented way | Yes |
Performance
On the V8 Web Tooling Benchmark at 4 vCPU / 8 GB, in StarSling's run of 3 October 2026 (run 37073680030, marked partial by StarSling), boat did 28.06 runs a second, second of 15, and Runloop 8.93, 15th of 15. Claude Managed Agents is not in that run, and Anthropic does not publish how many vCPUs a cloud sandbox gets, so we cannot compare it. 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. Providers ran on different hardware in different regions. A Managed Agents session often spends most of its time waiting on the model, not on the CPU.
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. 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. Claude Managed Agents 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 | Runloop, a month | Claude Managed Agents, 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 | $113 · Basic | $13 · Claude API (pay-as-you-go, usage tiers) · with limits | $20 · $20/mo |
| One agent that never stops 4 vCPU / 8 GB, 50 GB disk, running 24/7 all month | $429 · Basic | $53 · Claude API (pay-as-you-go, usage tiers) · 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 | $5,268 · Basic | $634 · Claude API (pay-as-you-go, usage tiers) · 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 | $1,441 · Basic | $360 · Claude API (pay-as-you-go, usage tiers) · 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 | $3,184 · Basic | $796 · Claude API (pay-as-you-go, usage tiers) · with limits | $8,800 · $2000/mo · with limits |
Runloop: Runloop bills per second for allocated CPU ($0.108 per vCPU-hour) and memory ($0.0252 per GB-hour) while a devbox starts, runs, suspends or resumes: $0.63 an hour at 4 vCPU / 8 GB. Its disk bills $0.2499 per GB-month for the devbox's whole life. Basic has no plan fee. Pro adds $250 a month that is not usage credit, and unlocks suspend and resume.
Claude Managed Agents: Managed Agents bills model tokens at API list rates plus $0.08 per session-hour of runtime, measured to the millisecond only while the session is running, including time spent waiting on the model. The tables show runtime only, and the cloud sandbox size is not published. Tokens come on top: in Anthropic's own example, a 1-hour Opus 5 session costs $0.625 in tokens and $0.08 in runtime. A self-hosted worker adds the cost of your own host.
Claude Managed Agents: Plans cost $20, $100, $500 or $2,000 a month. The whole plan price comes back as VM time for that month and does not carry over. A sandbox bills per second while it runs, busy or idle: $0.036 an hour at 4 vCPU / 8 GB. A stopped sandbox has no compute charge and keeps its disk for at least 14 days, but saved snapshots above the included count and the AI credentials proxy still charge. The $20 plan includes 12 starts a minute and 200 a day; extra starts cost $0.03 to $0.07 each. Model tokens for the built-in agents go to your model provider.
Limits in those scenarios
Runloop's keep-alive timer defaults to 1 hour and tops out at 48 hours unless an idle policy replaces it. Its disk tops out at 64 GiB, and it publishes no egress price or concurrency cap for paid plans. A Managed Agents cloud sandbox has up to 8 GB of memory and 10 GB of disk, an unpublished vCPU count and Docker only in limited availability, so no scenario here counts as a like-for-like fit; a self-hosted worker runs on a machine that you supply. Anthropic publishes 300 creates a minute per organization and usage-tier spend caps, not how many sessions can run at once. boat's limit that bites is start rate: the $2,000 plan includes 90 starts a minute, charges for more and refuses starts above 270 a minute, so 1,000 short jobs at once does not fit.
| Scenario | Runloop | Claude Managed Agents | boat |
|---|---|---|---|
| One coding agent, workdays |
|
| 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 |
| 1,000 short jobs at once |
|
|
|
Which one should you use?
Use Runloop when
you want your own choice of coding agents on micro-VM devboxes, large bursts of short jobs, arm64, Dockerfile builds, hostname allowlists or devboxes in your own AWS, GCP or Azure account.
- Any coding agent, on a micro-VM. Teams that want Claude Code, Codex or Gemini CLI side by side get a micro-VM devbox with Docker-in-Docker for each.
- Large waves of short jobs. Useful for RL rollouts and evals: Runloop publishes no start-rate limit on paid plans and sells RL fine-tuning on Enterprise.
- arm64, Dockerfiles and a way in. Runloop builds images from Dockerfiles, offers arm64, and opens an HTTPS tunnel and SSH into each devbox.
Use Claude Managed Agents when
you want Claude itself to do long tasks with no harness to build, idle time that costs no runtime, and the option to run tools on your own workers.
- No agent loop to build. Teams that want Claude doing long tasks get the harness, caching, compaction, memory, budgets and webhooks ready to use.
- Idle time costs no runtime. Agents that wait a long time between messages pay runtime only while working, and resume on the next message.
- Tools on your own workers. Teams with strict network rules can run every tool call on their own infrastructure while Anthropic runs the agent.
Where boat fits
If you are choosing between renting sandboxes and a hosted agent, boat sits closer to Runloop. Each agent gets a full VM with /dev/kvm, a disk saved every minute on every plan, and ports you can open. It runs Claude Code, Codex and five other agents that you drive from outside, so you are not tied to one harness.
A full machine with /dev/kvm
Runloop documents Docker-in-Docker but not nested virtualization. A Managed Agents cloud sandbox has up to 8 GB of memory, with Docker in limited availability; a self-hosted worker removes that, but you run the host. On boat, an Android emulator or a VM runs inside the sandbox.
Disk saved with no code
On Runloop Basic you take disk snapshots yourself, and Pro ($250 a month) adds idle suspend. Managed Agents keeps cloud sandbox files 30 days after creation. boat saves the disk every minute and at stop, and keeps a stopped disk for at least 14 days, or longer as a saved snapshot.
Open what the agent builds
A Managed Agents cloud sandbox has no documented inbound ports or SSH, and Runloop gives one HTTPS tunnel with no public IP. On boat, a dev server is reachable on HTTPS URLs for up to 50 ports, with SSH and raw ports on IPv6.
Use boat when your agents need a full machine for hours: Docker, emulators, a desktop and browser, ports you can open, a disk saved every minute, and the choice of Claude Code, Codex or other agents. Compare it one to one: boat vs Runloop and boat vs Claude Managed Agents.
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 Runloop cheaper than Claude Managed Agents?
For runtime alone, no. One coding agent on workdays is $113 a month on Runloop and $13 on Managed Agents, and 50 agents in parallel are $5,268 against $634. Managed Agents adds model tokens on top, and its cloud sandbox size is not published, so the machines are not like-for-like. boat is $20 and $317 for those two, plus your model tokens.
Which is faster, Runloop or Claude Managed Agents?
We cannot say. In StarSling's run of 3 October 2026 (run 37073680030), Runloop did 8.93 runs a second on the V8 Web Tooling Benchmark at 4 vCPU / 8 GB, and boat 28.06, on an AMD Ryzen 9 9950X host. Managed Agents is not in that run, and Anthropic does not publish the vCPU count of a cloud sandbox.
Can Runloop or Claude Managed Agents run Docker?
Runloop runs Docker as root on its Docker-in-Docker blueprints. The Managed Agents cloud sandbox lists Docker as limited availability; a self-hosted worker can run Docker if your host does. boat runs Docker with root and systemd in every sandbox, plus /dev/kvm for emulators and VMs.
Can I run Claude Code on Runloop or Claude Managed Agents?
On Runloop, yes: Claude Code, Codex, OpenCode and others run in a devbox through public agent wrappers. Managed Agents runs Anthropic's own harness with Claude models. boat has Claude Code, Codex and five other agents installed.
Can I open a web app my agent built?
On Runloop, yes, through one HTTPS tunnel per devbox. A Managed Agents cloud sandbox has no documented inbound ports or SSH; with a self-hosted worker you open ports on your own host. boat gives each sandbox HTTPS URLs on up to 50 ports, SSH and raw ports on IPv6.
Try boat
Install the CLI, then run boat new.
7-day free trial