Tokenomics: Quantifying Where Tokens Are Used in Agentic Software Engineering

TL;DR

This paper analyzes token consumption in LLM-based multi-agent software engineering, revealing that iterative review stages dominate costs and input tokens are the largest share, providing insights for optimizing workflows.

Contribution

It introduces a standardized framework for quantifying token usage across SDLC stages in LLM-MA systems, highlighting inefficiencies and guiding future optimization.

Findings

Code Review accounts for 59.4% of token consumption.

Input tokens make up 53.9% of total tokens used.

Token costs are higher during iterative refinement and verification.

Abstract

LLM-based Multi-Agent (LLM-MA) systems are increasingly applied to automate complex software engineering tasks such as requirements engineering, code generation, and testing. However, their operational efficiency and resource consumption remain poorly understood, hindering practical adoption due to unpredictable costs and environmental impact. To address this, we conduct an analysis of token consumption patterns in an LLM-MA system within the Software Development Life Cycle (SDLC), aiming to understand where tokens are consumed across distinct software engineering activities. We analyze execution traces from 30 software development tasks performed by the ChatDev framework using a GPT-5 reasoning model, mapping its internal phases to distinct development stages (Design, Coding, Code Completion, Code Review, Testing, and Documentation) to create a standardized evaluation framework. We then quantify and compare token distribution (input, output, reasoning) across these stages. Our preliminary findings show that the iterative Code Review stage accounts for the majority of token consumption for an average of 59.4% of tokens. Furthermore, we observe that input tokens consistently constitute the largest share of consumption for an average of 53.9%, providing empirical evidence for potentially significant inefficiencies in agentic collaboration. Our results suggest that the primary cost of agentic software engineering lies not in initial code generation but in automated refinement and verification. Our novel methodology can help practitioners predict expenses and optimize workflows, and it directs future research toward developing more token-efficient agent collaboration protocols.