Usage Logs
Usage logs track token consumption and cost per agent run. Use them to understand which agents and workflows drive the most LLM usage.
Viewing usage logs
Go to Settings → Logs → Usage in the Synapse UI.
What's tracked
| Metric | Description |
|---|---|
| Total tokens | Input tokens + output tokens per run |
| Input tokens | Tokens in the prompt (context + system prompt + history) |
| Output tokens | Tokens generated by the model |
| Cost (USD) | Estimated cost based on the model's pricing |
| Model | The LLM used for the run |
| Agent / Orchestration | What triggered the run |
| Session ID | Conversation session |
| Timestamp | When the run occurred |
Cost tracking in orchestrations
Orchestrations track cumulative cost across all steps:
{
"total_tokens_used": 15234,
"total_cost_usd": 0.0234
}
If max_total_cost_usd is set on the orchestration, execution halts when the cost exceeds the budget. The run transitions to failed with a cost-limit error.
Cost estimates
Cost estimates are calculated based on published pricing for each model. They are approximate — actual costs may differ due to caching, batch discounts, or pricing changes. Always verify with your LLM provider's billing dashboard.
Log retention
Usage logs are stored in DATA_DIR/logs/. They are not automatically rotated — for long-running deployments, periodically archive or delete old log files to prevent disk growth.