Microsoft Research and a team from the University of California, Berkeley have released TokenEconomizer, an open-source cost controller designed for coding agents. The plugin learns from prior conversations within a codebase and uses that history to reduce token waste in later sessions. It works with Claude Code, Codex, GitHub Copilot CLI, and Pi, and does so without changing model weights or modifying agent code. In tests covering 277 SWE-Bench tasks, the tool reduced average costs by 18.4% to 30.9% across five agent-model configurations. The release was cited by MarkTechPost.
Microsoft Research and a research team from the University of California, Berkeley have released TokenEconomizer, an open-source cost controller aimed at lowering token spending for coding agents.
Built to cut token waste from prior codebase sessions
The plugin learns from past conversations tied to a codebase and uses that information to reduce token waste in later sessions. It is designed for Claude Code, Codex, GitHub Copilot CLI, and Pi.
According to the release, TokenEconomizer does not alter model weights and does not require changes to agent code.
Results across 277 SWE-Bench tasks
In evaluations on 277 SWE-Bench tasks, the tool delivered average cost reductions ranging from 18.4% to 30.9% across five agent-model setups.
MarkTechPost was cited as the source of the report.
This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan. Disclaimer:
The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.
Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.
