# Token Monitoring

Monitor and optimize token usage across all your AI operations.

## Token Tracking

Track every token consumed by your AI systems:

```typescript
// Get token usage
const usage = await ants.finops.getTokenUsage({
  period: 'last_24h',
  groupBy: 'model'
})

console.log(`GPT-4: ${usage.gpt4} tokens`)
console.log(`GPT-3.5: ${usage.gpt35} tokens`)
console.log(`Total cost: $${usage.totalCost}`)
```

## Usage by Model


<table><thead><tr><th>Model</th><th>Cost per 1K tokens</th><th>Use Case</th></tr></thead><tbody><tr><td>GPT-4</td><td>$0.03</td><td>Complex reasoning</td></tr><tr><td>GPT-3.5 Turbo</td><td>$0.002</td><td>Simple tasks</td></tr><tr><td>Claude 3</td><td>$0.015</td><td>Long context</td></tr></tbody></table>


## Optimization Tips

1. Use smaller models where appropriate
2. Cache common responses
3. Optimize prompts for brevity
4. Implement token limits

## Next Steps

- [Cost Tracking](/docs/aigovernance/aicost/cost-tracking)
- [Budget Management](/docs/aigovernance/aicost/budgets)

Model	Cost per 1K tokens	Use Case
GPT-4	$0.03	Complex reasoning
GPT-3.5 Turbo	$0.002	Simple tasks
Claude 3	$0.015	Long context