BurnRate

Local-first observability for AI coding tools

BurnRate provides a free local-first observability tool for various AI coding providers, allowing users to track costs, observe subagent behavior, and optimize spending directly from their terminal. This feature ensures that no data is sent to the cloud, maintaining user privacy.

Subagent observability

This feature allows users to see every subagent spawned by Claude Code, including their costs, tool usage, and duration. It provides a detailed cost breakdown for each subagent, enabling users to understand where their tokens are being consumed.

Multi-provider analytics

BurnRate enables users to track and compare costs across multiple AI coding tools, including Claude Code, Cursor, Codex, and others. This feature provides insights into cost efficiency and usage patterns across all supported providers.

Cost Breakdown & Plan Value

This feature allows users to see the cost of their AI usage compared to API pricing, providing insights into daily spending trends and model-specific cost distributions. It helps users understand the value of their subscription versus API costs.

Budget Alerts

Users can set spending caps and receive alerts when their budgets are at risk. This feature includes desktop notifications and Slack webhooks for team-wide alerts, ensuring that users stay informed about their spending.

Rate Limit Monitoring

This feature tracks how close users are to hitting rate limits across various AI providers. It provides live utilization bars with color-coded thresholds and reset countdowns, helping users avoid unexpected throttling.

Optimization Engine

BurnRate's optimization engine analyzes usage patterns and suggests actionable savings through 46 provider-specific rules. Each rule includes an impact score and estimated monthly savings, helping users optimize their AI tool usage effectively.

Model Recommendations

This feature provides AI-powered suggestions for which model fits each project best, helping users avoid overpaying for simple tasks. It analyzes usage patterns to recommend more cost-effective models.

Operation Cost Breakdown

Users can see which operations, such as reading or writing, consume the most tokens. This feature helps identify areas for optimization and cost reduction by analyzing operation-level token usage.

Session Explorer

This feature allows users to browse every AI session with full metadata, including duration, tokens, cost, model, and project. Users can filter and sort sessions to find their most expensive interactions.

Provider Comparison

BurnRate provides a side-by-side comparison of costs across all supported AI providers. Users can analyze per-provider model breakdowns, efficiency scores, and daily cost trends to make informed decisions.

Spend Forecasting

This feature predicts future spending based on historical usage patterns, providing users with insights into projected monthly and weekly costs. It helps users plan their budgets effectively.