SWE-Lancer: Can Frontier LLMs Earn $1 Million from Real-World Freelance Software Engineering?

TL;DR

SWE-Lancer is a comprehensive benchmark of over 1,400 real-world freelance software engineering tasks valued at $1 million, designed to evaluate the capabilities and economic potential of frontier language models in practical software engineering scenarios.

Contribution

This paper introduces SWE-Lancer, the first large-scale, real-world software engineering benchmark with monetary valuation, including both technical and managerial tasks, and provides an open-source evaluation platform.

Findings

Frontier models struggle to solve most tasks.

Benchmark covers diverse engineering and managerial tasks.

Open-source tools facilitate future AI research in software engineering.

Abstract

We introduce SWE-Lancer, a benchmark of over 1,400 freelance software engineering tasks from Upwork, valued at \$1 million USD total in real-world payouts. SWE-Lancer encompasses both independent engineering tasks--ranging from \$50 bug fixes to \$32,000 feature implementations--and managerial tasks, where models choose between technical implementation proposals. Independent tasks are graded with end-to-end tests triple-verified by experienced software engineers, while managerial decisions are assessed against the choices of the original hired engineering managers. We evaluate model performance and find that frontier models are still unable to solve the majority of tasks. To facilitate future research, we open-source a unified Docker image and a public evaluation split, SWE-Lancer Diamond (https://github.com/openai/SWELancer-Benchmark). By mapping model performance to monetary value, we hope SWE-Lancer enables greater research into the economic impact of AI model development.