traceburn is an open-source, local-first tracer and efficiency profiler designed for AI agent workflows. It enables developers to analyze cost and latency using flamegraphs, perform deterministic replay, compare runs, and detect inefficiencies, all stored in a single SQLite file. Ideal for AI developers seeking detailed observability and optimization tools.
In the LLM eval & observability space, traceburn takes a focused approach. It focuses on profiling and analyzing the efficiency and cost of AI agent workflows locally. It is built as an open-source project for AI developers and researchers. The project is open source (MIT). It ships for the command line, and it can be self-hosted.
Behind traceburn is TommyTranX, and it first shipped in 2026. Development happens publicly on GitHub with 10 commits in the last 90 days. Among its 6 catalogued features are cost profiling, latency flamegraphs, and deterministic replay.
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