Cua Fleets vs Tensorlake Sandboxes
Cua Fleets costs $0.044625 per vCPU-hour and Tensorlake Sandboxes costs $0.07 per vCPU-hour. Cua Fleets runs gVisor. Tensorlake Sandboxes runs microVMs, offers GPUs and has no session limit.
Updated
Tensorlake Sandboxes
- Isolation
- MicroVM
- Free tier
- Free plan, $0 forever
- Max session
- No limit
What is different
8 fields differ. The rest are the same or not listed.
Cua Fleets
Tensorlake Sandboxes
- vCPU, per hour
- Cua Fleets: $0.044625
- Tensorlake Sandboxes: $0.07
- RAM, per GiB-hour
- Cua Fleets: $0.0223125
- Tensorlake Sandboxes: $0.015
- Storage
- Cua Fleets: Free
- Tensorlake Sandboxes: $0.0002 per GiB-hour
- Billing
- Cua Fleets: Per second
- Tensorlake Sandboxes: Active CPU time only
- Largest sandbox
- Cua Fleets: 8 vCPU · 32 GiB
- Tensorlake Sandboxes: 16 vCPU · 64 GiB
- Isolation
- Cua Fleets: gVisor
- Tensorlake Sandboxes: MicroVM
- Pause and resume
- Cua Fleets: No
- Tensorlake Sandboxes:
- SDKs
- Cua Fleets: Python, TypeScript, Rust
- Tensorlake Sandboxes: Python, TypeScript
Full comparison
18 fields, at each provider's list price
Cua Fleetscua
Tensorlake Sandboxestensorlake
Pricing
- vCPU, per hour
- Cua Fleets: $0.044625
- Tensorlake Sandboxes: $0.07
- RAM, per GiB-hour
- Cua Fleets: $0.0223125
- Tensorlake Sandboxes: $0.015
- Storage
- Cua Fleets: Free
- Tensorlake Sandboxes: $0.0002 per GiB-hour
- Free tier
- Cua Fleets: Not listed
- Tensorlake Sandboxes: Free plan, $0 forever
- Paid plan from
- Cua Fleets: Not listed
- Tensorlake Sandboxes: Pay as you go
- Billing
- Cua Fleets: Per second
- Tensorlake Sandboxes: Active CPU time only
Limits
- Max session
- Cua Fleets: Not listed
- Tensorlake Sandboxes: No limit
- Idle stop
- Cua Fleets: Not listed
- Tensorlake Sandboxes: After 10 minutes
- Largest sandbox
- Cua Fleets: 8 vCPU · 32 GiB
- Tensorlake Sandboxes: 16 vCPU · 64 GiB
- Concurrent sandboxes
- Cua Fleets: Not listed
- Tensorlake Sandboxes: Up to 1,000
Platform
- Isolation
- Cua Fleets: gVisor
- Tensorlake Sandboxes: MicroVM
- GPU
- Cua Fleets: Not listed
- Tensorlake Sandboxes:
- Snapshots
- Cua Fleets: Not listed
- Tensorlake Sandboxes:
- Pause and resume
- Cua Fleets: No
- Tensorlake Sandboxes:
Developer
- SDKs
- Cua Fleets: Python, TypeScript, Rust
- Tensorlake Sandboxes: Python, TypeScript
- Integrations
- Cua Fleets: OpenAI Agents SDK, MCP
- Tensorlake Sandboxes: Not listed
Trust
- License
- Cua Fleets: Not listed
- Tensorlake Sandboxes: Not listed
- BYOC
- Cua Fleets: Your AWS, Google Cloud, or Modal account
- Tensorlake Sandboxes: Your AWS, GCP or Azure account
Hourly cost
One sandbox running for one hour
Cua Fleets
Tensorlake Sandboxes
- 1 vCPU · 2 GiB
- Cua Fleets: $0.089
- Tensorlake Sandboxes: $0.100
- 2 vCPU · 4 GiB
- Cua Fleets: $0.179
- Tensorlake Sandboxes: $0.200
- 4 vCPU · 8 GiB
- Cua Fleets: $0.357
- Tensorlake Sandboxes: $0.400
Compute only, at list rates.
Open the calculatorWhich one fits
Based on the fields above
Cua Fleets fits when
- You use the OpenAI Agents SDK or MCP.
- You want the lower list price: $0.179 against $0.200 an hour for 2 vCPU and 4 GiB.
- You need SOC 2.
Tensorlake Sandboxes fits when
- You need GPUs.
- You want each sandbox in its own microVM.
- You need snapshots.
- You need pause and resume.
More comparisons
Agent Sandbox vs Cua Fleets- Agent Sandbox vs Tensorlake Sandboxes
Agent37 Cloud vs Cua Fleets- Agent37 Cloud vs Tensorlake Sandboxes
AgentBay vs Cua Fleets- AgentBay vs Tensorlake Sandboxes
- Alibaba Cloud Agent Sandbox vs Cua Fleets
- Alibaba Cloud Agent Sandbox vs Tensorlake Sandboxes