Insights
View insights into token usage for your models
Insights provides detailed analytics on token usage patterns, helping you understand consumption habits across different models. You can also view cost estimates and usage breakdowns by specific models, enabling more informed decision-making about AI tool usage.
Costs and token usage are estimations only. Please refer your model provider account for accurate token usage and costs.
Summary
The Summary tab provides an overview of token usage and cost estimates. You can even filter by date range to see how usage has changed over time - including a temperature map visualization of day-by-day usage.
Provider
The Provider tab provides more detailed insights into token usage by Provider. You can further break down usage by individual models, allowing you to identify which models are consuming the most tokens. This helps in optimizing costs and understanding model-specific usage patterns.
Also provided are graphs displaying model throughput and latency, which can help you identify any performance issues or bottlenecks in model usage.