ABC-Bench: Benchmarking Agentic Backend Coding in Real-World Development

TL;DR

ABC-Bench is a new benchmark that evaluates AI agents on realistic backend development tasks, including environment setup and deployment, revealing current models' limitations in practical software engineering scenarios.

Contribution

We introduce ABC-Bench, the first benchmark focusing on agentic backend coding in real-world workflows, covering end-to-end development from exploration to deployment.

Findings

State-of-the-art models perform poorly on holistic backend tasks.

Current models struggle with environment configuration and service deployment.

ABC-Bench highlights the gap between AI capabilities and practical engineering requirements.

Abstract

The evolution of Large Language Models (LLMs) into autonomous agents has expanded the scope of AI coding from localized code generation to complex, repository-level, and execution-driven problem solving. However, current benchmarks predominantly evaluate code logic in static contexts, neglecting the dynamic, full-process requirements of real-world engineering, particularly in backend development which demands rigorous environment configuration and service deployment. To address this gap, we introduce ABC-Bench, a benchmark explicitly designed to evaluate agentic backend coding within a realistic, executable workflow. Using a scalable automated pipeline, we curated 224 practical tasks spanning 8 languages and 19 frameworks from open-source repositories. Distinct from previous evaluations, ABC-Bench require the agents to manage the entire development lifecycle from repository exploration to instantiating containerized services and pass the external end-to-end API tests. Our extensive evaluation reveals that even state-of-the-art models struggle to deliver reliable performance on these holistic tasks, highlighting a substantial disparity between current model capabilities and the demands of practical backend engineering. Our code is available at https://github.com/OpenMOSS/ABC-Bench.