Skip to main content

​
Overview

RunTools enforces limits to ensure fair usage and platform stability. Limits vary by plan — see Pricing for plan details.

​
Sandbox Limits

LimitFreeProTeamEnterprise
Concurrent sandboxes520100Unlimited
Max sandbox runtime1 hour4 hours24 hoursUnlimited
Idle timeout (auto-pause)10 min30 min2 hoursCustom
vCPUs per sandbox24816
Memory per sandbox2 GB4 GB16 GB64 GB
Disk per sandbox10 GB50 GB100 GB500 GB
Snapshots per sandbox31050Unlimited
Snapshot retention7 days30 days90 daysUnlimited

​
Storage Limits

LimitFreeProTeamEnterprise
Total workspace storage10 GB50 GB500 GBUnlimited
File upload size100 MB500 MB2 GB10 GB
Files per directory10,00050,000100,000Unlimited

​
Agent Limits

LimitFreeProTeamEnterprise
Agent runtimes31050Unlimited
Agent deployments31050Unlimited
Concurrent runs520100Unlimited
Run timeout5 min30 min2 hoursCustom
Max tokens per run100K200K500KUnlimited
Schedules-10100Unlimited
Triggers-10100Unlimited

​
Tool Limits

LimitFreeProTeamEnterprise
Tool calls/month10010,000100,000Unlimited
Custom tools320UnlimitedUnlimited
Tool execution timeout30s60s5 min30 min

​
Code Execution Limits

LimitFreeProTeamEnterprise
Executions/month1005,00050,000Unlimited
Execution timeout30s60s5 min30 min
Output size1 MB10 MB50 MB100 MB

​
AIP Hub Limits

LimitFreeProTeamEnterprise
Sessions101001,000Unlimited
Blocks per session1001,00010,000Unlimited
Total context blocks1,00010,000100,000Unlimited
Max tokens per session100K500K2MUnlimited
Search queries/day1001,00010,000Unlimited

​
API Rate Limits

LimitFreeProTeamEnterprise
Requests/minute603001,000Custom
Requests/hour1,00010,00050,000Custom
WebSocket connections520100Unlimited
Burst limit1050200Custom

​
Rate Limit Headers

Every API response includes rate limit headers: 
X-RateLimit-Limit: 300
X-RateLimit-Remaining: 299
X-RateLimit-Reset: 1609459200

​
Handling Rate Limits

import { RateLimitError } from '@runtools/sdk';

try {
  await rt.sandboxes.create({ template: 'nodejs-20' });
} catch (error) {
  if (error instanceof RateLimitError) {
    console.log(`Rate limited. Retry after ${error.retryAfter}ms`);
    await new Promise(r => setTimeout(r, error.retryAfter));
    // Retry...
  }
}

​
Checking Usage

const usage = await rt.usage.get();

console.log(usage);
// {
//   sandboxes: { active: 3, limit: 20 },
//   storage: { usedGb: 12.5, limitGb: 50 },
//   toolCalls: { thisMonth: 4521, limit: 10000 },
//   executions: { thisMonth: 892, limit: 5000 },
//   apiCalls: { thisMinute: 45, limit: 300 },
// }

// Check specific resource
const canCreate = usage.sandboxes.active < usage.sandboxes.limit;

​
Limit Notifications

Get notified when approaching limits: 
// Configure notifications
await rt.notifications.configure({
  usageWarning: 80,  // Notify at 80% usage
  channels: ['email', 'webhook'],
  webhookUrl: 'https://myapp.com/runtools-alerts',
});

​
Requesting Limit Increases

For temporary or permanent limit increases: 
// Request limit increase
await rt.limits.requestIncrease({
  resource: 'concurrent_sandboxes',
  requested: 50,
  reason: 'Load testing for product launch',
  duration: '7d',  // Temporary increase
});
Or contact support for Enterprise-level custom limits.

​
Best Practices

Paused sandboxes don’t count against concurrent limits. Pause when not in use.
Instead of listing 50 tools in your agent, use tool_search for dynamic discovery.
When rate limited, use exponential backoff with jitter for retries.
Set up notifications at 80% to avoid hitting limits unexpectedly.
Delete snapshots you no longer need to free up storage.