Cloud sandbox for AI agents with 94% lower cost than most providers
Deploy any open model from Ollama or Hugging Face as a private, OpenAI-compatible API
Stage
Not given
Not given
Founded
Not given
Not given
Based in
Not given
US
Pricing model
Paid
Paid
Pricing
Standard sandbox: $0.80 per vCPU, $0.70 per GB RAM, $0.09 per GB disk per 730-hour month, metered by the minute. Example: 2 vCPU 4 GB with 4 GB disk costs $4.76/month if running continuously. Idle disk alone costs $0.36/month for 4 GB. VM sandbox costs 2.5x compute rate. New accounts get $5 starter
Pay as you go: $0/month with prepaid credits from $10. Pro: $29/month with $29 monthly credits. Business: $299/month with $299 monthly credits. GPU pricing: NVIDIA L4 $0.95/hour ($0.000264/second, $1,180/month always-on); NVIDIA L40S $3.20/hour ($0.000889/second, $3,974/month always-on); 4x NVIDIA L
Key features
Per-minute billing for compute, disk storage metered monthly
Persistent file storage with automatic backups
Auto-sleep to pause compute charges when idle
Kernel-enforced isolation via gVisor
Docker container support
Outbound internet access only, firewalled per sandbox
Scheduled runs with cron
Support for multiple AI models and custom keys
Public HTTPS URLs and custom domain support
Deploy models with one command via CLI
OpenAI and Ollama API compatibility
Scales to zero with configurable idle timeout (60 seconds to 60 minutes)
Private endpoints with no prompt logging
Supports GGUF and safetensors model formats
Free model storage (10GB to 1TB depending on plan)
Always-on option for no cold starts
Batch jobs pricing 30-50% cheaper
What makes it different
94% lower cost than most sandbox providers
No subscription fees, plan fees, or minimum commitments
Compute billing stops when sandbox sleeps or stops
Data never collected for training, SOC 2 Type I compliant
Agents can run indefinitely without timeouts
No Dockerfile or server management required
Use your existing OpenAI or Ollama client with one URL change
Pay only for actual compute seconds used, not monthly subscription
Cold start time never billed and requests wait instead of failing
Prompts and outputs never logged or used for training
Agent 37
BitsRun