Anthropic Code Execution vs Cua Fleets

Anthropic Code Execution costs $0.05 per vCPU-hour. Anthropic Code Execution runs containers. Cua Fleets runs gVisor.

Updated Oct 15, 2026

Anthropic Code Execution

Sandboxed Python and bash containers that Claude runs through the Anthropic API.

Isolation: Container
Free tier: 1,550 free hours per month
Max session: Not listed

Visit site Anthropic Code Execution profile

Cua Fleets

Hosted Linux and Windows sandboxes for computer-use agents.

Isolation: gVisor
Free tier: Not listed
Max session: Not listed

Visit site Cua Fleets profile

What is different

4 fields differ. The rest are the same or not listed.

Field

Anthropic Code Execution

Cua Fleets

Billing: Anthropic Code Execution: Bundled machine time; Cua Fleets: Per second
Isolation: Anthropic Code Execution: Container; Cua Fleets: gVisor
Pause and resume: Anthropic Code Execution:; Cua Fleets: No
SDKs: Anthropic Code Execution: Python, TypeScript, Go, Java, Ruby; Cua Fleets: Python, TypeScript, Rust

Full comparison

18 fields, at each provider's list price

Anthropic Code Executionplatform.claude.com/docs/en/about-claude/pricingplatform.claude.com

Cua Fleetscua.ai/#pricing

Pricing

vCPU, per hour: Anthropic Code Execution: $0.05; Cua Fleets: Not listed
RAM, per GiB-hour: Anthropic Code Execution: Included; Cua Fleets: Not listed
Storage: Anthropic Code Execution: Not listed; Cua Fleets: Not listed
Free tier: Anthropic Code Execution: 1,550 free hours per month; Cua Fleets: Not listed
Paid plan from: Anthropic Code Execution: Not listed; Cua Fleets: Not listed
Billing: Anthropic Code Execution: Bundled machine time; Cua Fleets: Per second

Limits

Max session: Anthropic Code Execution: Not listed; Cua Fleets: Not listed
Idle stop: Anthropic Code Execution: Not listed; Cua Fleets: Not listed
Largest sandbox: Anthropic Code Execution: 1 vCPU · 5 GiB; Cua Fleets: Not listed
Concurrent sandboxes: Anthropic Code Execution: Not listed; Cua Fleets: Not listed

Platform

Isolation: Anthropic Code Execution: Container; Cua Fleets: gVisor
GPU: Anthropic Code Execution: Not listed; Cua Fleets: Not listed
Snapshots: Anthropic Code Execution: Not listed; Cua Fleets: Not listed
Pause and resume: Anthropic Code Execution:; Cua Fleets: No

Developer

SDKs: Anthropic Code Execution: Python, TypeScript, Go, Java, Ruby; Cua Fleets: Python, TypeScript, Rust
Integrations: Anthropic Code Execution: Not listed; Cua Fleets: OpenAI Agents SDK, MCP, Claude Code

Trust

License: Anthropic Code Execution: Not listed; Cua Fleets: Not listed
BYOC: Anthropic Code Execution: Not listed; Cua Fleets: Your AWS, Google Cloud, or Modal account

Hourly cost

One sandbox running for one hour

Size

Anthropic Code Execution

Cua Fleets

1 vCPU · 2 GiB: Anthropic Code Execution: $0.050; Cua Fleets: Not listed
2 vCPU · 4 GiB: Anthropic Code Execution: $0.100; Cua Fleets: Not listed
4 vCPU · 8 GiB: Anthropic Code Execution: $0.200; Cua Fleets: Not listed

Compute only, at list rates.

Open the calculator

Which one fits

Based on the fields above

Anthropic Code Execution fits when

You need pause and resume.

Cua Fleets fits when

You use the OpenAI Agents SDK, MCP or Claude Code.
You need a desktop in the sandbox.
You need the sandboxes in your own cloud account.
You need SOC 2.

More comparisons

See all comparisons

Anthropic Code Execution

Cua Fleets

What is different

Full comparison

Pricing

Limits

Platform

Developer

Trust

Hourly cost

Which one fits

Anthropic Code Execution fits when

Cua Fleets fits when

More comparisons

Related

Compare

Pricing

Browse