Last updated: 10/5/2026Last verified: 2026-10-02
What is Battleships?
Battleships is a pricing engine and directory for comparing cloud sandbox, virtual machine, and agent-runtime providers for AI-agent workloads. It estimates monthly costs using provider billing models and links pricing and feature answers to quoted public sources. The directory is designed for both human research and programmatic use by coding agents.
Key Features
- Compares 366 sandbox, VM, and agent-runtime providers
- Models billing details such as per-second usage, minimum charges, idle time, starts, plans, and seats
- Interactive estimator with presets for fleet size, machine shape, usage hours, storage, network reach, browser or desktop access, and compliance requirements
- Links every price and feature answer to a quoted source
- Keeps unpublished figures marked as unknown instead of assuming they are free or unlimited
- Provides agent-readable static files, including llms.txt, provider JSON cards, rankings, and pricing-engine data
Best For
Pricing
Battleships is listed as free. The service compares provider costs rather than replacing each provider's own billing, so estimated totals should be checked against the linked source pages and the provider's current terms.
Pros & Cons
Pros
- Focuses on workload-based cost estimation rather than simple headline price comparisons
- Accounts for billing details that can materially affect monthly infrastructure costs
- Provides quoted source links for pricing and feature answers
- Clearly distinguishes unpublished information from confirmed public data
- Offers structured files that can be read by coding agents and other automated workflows
- Covers a broad directory of sandbox, VM, and agent-runtime providers
Cons
- Estimated totals depend on the accuracy and currency of the linked public pricing information
- The directory does not make unpublished provider figures appear comparable or complete
- Users still need to validate assumptions such as usage, storage, network access, and seats
- Battleships is a comparison and estimation tool, not a replacement for provisioning or managing the selected infrastructure
Alternatives
Modal is an alternative for developers evaluating cloud execution infrastructure for AI and code workloads, while Battleships is focused on comparing multiple providers.
E2B provides cloud sandboxes for AI agents and is a relevant alternative when the goal is to use a dedicated agent-sandbox platform rather than compare providers.
Daytona is an alternative for AI development environments and sandboxed coding workflows, whereas Battleships centers on cross-provider pricing research.
Replit offers an AI-oriented development environment and coding workflow that may suit users seeking an integrated product instead of a provider directory and estimator.
FAQ
Details
Platform
Features
- Prices your agent workload the way each of 366 sandbox, VM, and runtime providers bills it — per second, minimums, idle time, starts, plans, and seats
- Interactive estimator with use-case presets covering fleet size, machine shape, usage hours, storage, network reach, browser/desktop access, and compliance requirements
- Every price and feature answer links to its quoted source; unpublished numbers stay unknown instead of being assumed free or unlimited
- Agent-readable static files (llms.txt, provider JSON cards, rankings, pricing engine) so coding agents can run the comparison programmatically
Languages
Known limitations
- Covers cloud sandbox, VM, and agent-runtime compute providers; general IaaS, dedicated bare-metal, and model API pricing are out of scope
- Reflects public list prices only — private deals, negotiated discounts, and billing edge cases are not included
- English-only interface; provider prices and features were last re-checked by the maintainers around 2026-09-28
- The source repository publishes data and methodology without a declared license
