BurnRate is a local-first analytics tool designed for developers and teams to monitor, analyze, and optimize the costs and behavior of AI coding agents. It provides observability for AI tools such as Claude Code, Cursor, Copilot, Codex, Windsurf, Cline, and Aider, supporting a total of seven providers and over forty AI models. The platform enables users to see detailed breakdowns of AI agent activity and spending, offering insights into both overall and subagent-level costs.
The tool features nine analytics views, including cost breakdowns, plan value comparisons, daily spending trends, model-specific cost analysis, and session explorers that display metadata such as duration, tokens, cost, and project. Users can filter and sort session histories, compare providers side-by-side, and view efficiency scores and token distributions. BurnRate also includes a project analytics suite to track costs and usage per project, set per-project budgets, and monitor budget caps. The optimization engine applies 46 rules to suggest actionable savings, providing provider-specific configuration snippets and estimating monthly savings. Additionally, the tool offers operation-level breakdowns, context efficiency metrics, and rate limit monitoring with live utilization bars and reset countdowns.
BurnRate is delivered as a desktop and terminal-based application that runs entirely on the user's machine, ensuring that code and session data remain private and never leave the local environment. It requires no configuration, account, or cloud dependency, and can be installed via Homebrew or direct binary download. For teams, it supports desktop notifications and Slack webhooks for budget alerts, and offers both individual and team-wide cost management features.
The service is completely free for individual developers with no signup required, and provides multiple pricing tiers for teams and enterprises, including Free, Pro, Team, and Enterprise plans. BurnRate positions itself within the AI agent observability and cost analytics class, specifically targeting developers and organizations seeking to optimize their use of AI coding tools and manage associated expenses.
In the LLM eval & observability space, BurnRate takes a focused approach. Helping developers track, analyze, and optimize the costs and behaviors of AI coding agents across multiple providers. It is built as a B2B product for developers using AI coding tools. There is a free tier. It runs on the command line, macOS, Windows, and Linux.
BurnRate first shipped in 2026. Key capabilities include cost analytics, agent observability, and subagent tracking.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do