aifollow.news
Back Rohan Paul
Rohan Paul· @rohanpaul_ai · X· · Original publication time AI score54

TokenEconomizer study uses an external controller to cut coding-agent token costsMachine translation

AI introduction

A UC Berkeley and Microsoft paper proposes an external controller that learns from past coding-agent runs to inject useful facts, remove unused tools, and stop failing loops early. Across 277 SWE-Bench tasks using Claude Code, Codex, and GitHub Copilot CLI, the reported mean per-task cost fell by 18.4% to 30.9%.

Original post · Original

The article text is unavailable in this language; an existing version is shown.

@rohanpaul_ai

– https://t.co/rBBRuMoum2

Title: "TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents"

ReplyRohan Paul@rohanpaul_ai
New UC Berkeley and Microsoft paper cuts coding-agent token costs by 18.4% to 30.9% using an external controller that learns from past runs. by injecting learned facts, pruning unused tools, and stopping failed loops early. Coding agents are billed by the token, and each turn resends the growing history, so long projects get expensive. Token efficiency is usually measured on isolated tasks, missing the cost that builds up across a sequence of tasks in the same codebase. TokenEconomizer sits outside the agent as a black box, changing only prompts and tool settings. From past runs it learns which facts to inject, which unused tools to drop, and when to stop a failing loop. Across Claude Code, Codex, and GitHub Copilot CLI on 277 SWE-Bench tasks, it cut mean per-task cost 18.4% to 30.9%. It improved Pass@1 in 4 of 5 agent-model settings, and savings held in every repository. Measure agent cost across a whole project, not per task, and use an external controller that adapts to each repo, not fixed plugins.
View this post on X ↗

Source: Rohan Paul · x.com

Research
Found an error? Send a correction