modelmri is an open-source developer tool that provides Chrome DevTools-like debugging and interpretability features for AI models and agents running locally. It allows users to visualize attention, inspect features, and trace agent steps, helping AI developers and researchers better understand and debug their models.
modelmri is a Debugging & profiling project. It focuses on debugging and interpreting the internal workings of local AI models and agents. It is built as an open-source project for AI developers and researchers. modelmri is open source under the MIT license. It runs on the command line.
Behind modelmri is Muhammad Mahad Azher, and it first shipped in 2026. The project is developed in the open on GitHub with 2 commits in the last 90 days. Among its 5 catalogued features are attention visualization, agent step tracing, and feature inspection. modelmri is currently in beta.
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