Overview
Monitor your token consumption and associated costs across all instances.
Get Token Usage
Retrieve your usage statistics:
const usage = await client.usage.getTokenUsage();
console.log(`Tokens this month: ${usage.monthlyTokens}`);
console.log(`Cost this month: $${usage.monthlyCost}`);
console.log(`Total tokens: ${usage.totalTokens}`);
console.log(`Total cost: $${usage.totalCost}`);
console.log(`Last usage: ${usage.lastUsage}`);
Response
{
monthlyTokens: 1250000,
totalTokens: 5000000,
monthlyCost: 15.75,
totalCost: 63.00,
lastUsage: "2025-01-06T12:34:56Z"
}
Fields
| Field | Type | Description |
|---|
monthlyTokens | number | Tokens used this month |
totalTokens | number | Total tokens used (all time) |
monthlyCost | number | Cost this month in USD |
totalCost | number | Total cost in USD |
lastUsage | string | Timestamp of last token usage |
Track Task Usage
Each task execution returns usage data:
const result = await instance.run('Build a REST API');
console.log(`Tokens used: ${result.tokensUsed}`);
console.log(`Cost: $${result.usage.cost}`);
console.log(`Duration: ${result.duration_ms}ms`);
Task Response
{
response: "I've created a REST API...",
tokensUsed: 15432,
usage: {
total_tokens: 15432,
cost: 0.023
},
duration_ms: 45230
}
Get General Stats
Get general statistics about your account:
const stats = await client.usage.getStats();
console.log(stats);
This is a legacy endpoint maintained for backwards compatibility
Usage Monitoring Example
Track usage across multiple tasks:
const instance = await client.instances.create({ name: 'tracker' });
let totalTokens = 0;
let totalCost = 0;
// Task 1
const result1 = await instance.run('Create user model');
totalTokens += result1.tokensUsed;
totalCost += result1.usage.cost;
// Task 2
const result2 = await instance.run('Add authentication');
totalTokens += result2.tokensUsed;
totalCost += result2.usage.cost;
console.log(`Session tokens: ${totalTokens}`);
console.log(`Session cost: $${totalCost.toFixed(4)}`);
// Check monthly totals
const usage = await client.usage.getTokenUsage();
console.log(`Monthly total: ${usage.monthlyTokens} tokens`);
Cost Optimization Tips
Reuse Instances
Maintain conversation context across tasks instead of creating new instances
Smaller Models
Use appropriate model sizes for simpler tasks
Timeouts
Set reasonable timeouts to avoid runaway costs
Monitor Usage
Regularly check usage to detect anomalies
Set Usage Alerts
Monitor usage programmatically:
class UsageMonitor {
constructor(client, threshold) {
this.client = client;
this.threshold = threshold;
}
async check() {
const usage = await this.client.usage.getTokenUsage();
if (usage.monthlyCost > this.threshold) {
console.warn(`⚠️ Monthly cost ($${usage.monthlyCost}) exceeded threshold ($${this.threshold})`);
return false;
}
return true;
}
}
// Usage
const monitor = new UsageMonitor(client, 100); // $100 threshold
if (await monitor.check()) {
const instance = await client.instances.create({ name: 'my-agent' });
await instance.run('Build something');
}
Token usage: Charged per token consumedModel selection: Different models have different ratesInstance uptime: VM running timeStorage: Persistent storage costs
Check the in10nt dashboard for current pricing details
API Endpoints
Token Usage
Returns monthly and total token usage with costs.
General Stats
Returns general account statistics (legacy).
