FastCode: Fast and Cost-Efficient Code Understanding and Reasoning

TL;DR

FastCode introduces a novel framework that improves large-scale code reasoning by efficiently navigating code repositories with a structural scouting mechanism, reducing computational costs while maintaining high accuracy.

Contribution

FastCode presents a structure-aware, cost-efficient approach to code understanding that decouples exploration from content consumption, outperforming existing methods in accuracy and resource usage.

Findings

Outperforms state-of-the-art baselines in reasoning accuracy

Reduces token consumption significantly

Validated on multiple large-scale benchmarks

Abstract

Repository-scale code reasoning is a cornerstone of modern AI-assisted software engineering, enabling Large Language Models (LLMs) to handle complex workflows from program comprehension to complex debugging. However, balancing accuracy with context cost remains a significant bottleneck, as existing agentic approaches often waste computational resources through inefficient, iterative full-text exploration. To address this, we introduce FastCode, a framework that decouples repository exploration from content consumption. FastCode utilizes a structural scouting mechanism to navigate a lightweight semantic-structural map of the codebase, allowing the system to trace dependencies and pinpoint relevant targets without the overhead of full-text ingestion. By leveraging structure-aware navigation tools regulated by a cost-aware policy, the framework constructs high-value contexts in a single, optimized step. Extensive evaluations on the SWE-QA, LongCodeQA, LOC-BENCH, and GitTaskBench benchmarks demonstrate that FastCode consistently outperforms state-of-the-art baselines in reasoning accuracy while significantly reducing token consumption, validating the efficiency of scouting-first strategies for large-scale code reasoning. Source code is available at https://github.com/HKUDS/FastCode.