DeepCode: Open Agentic Coding

TL;DR

DeepCode introduces a novel autonomous framework that manages information flow to improve scientific paper-to-code synthesis, outperforming existing agents and human experts in generating high-quality code from research papers.

Contribution

It presents a new information-flow management approach for LLM-based code synthesis, addressing context limitations and achieving state-of-the-art results on scientific paper benchmarks.

Findings

Outperforms commercial agents like Cursor and Claude Code.

Surpasses PhD-level human experts on key metrics.

Transforms paper specifications into high-quality code autonomously.

Abstract

Recent advances in large language models (LLMs) have given rise to powerful coding agents, making it possible for code assistants to evolve into code engineers. However, existing methods still face significant challenges in achieving high-fidelity document-to-codebase synthesis--such as scientific papers to code--primarily due to a fundamental conflict between information overload and the context bottlenecks of LLMs. In this work, we introduce DeepCode, a fully autonomous framework that fundamentally addresses this challenge through principled information-flow management. By treating repository synthesis as a channel optimization problem, DeepCode seamlessly orchestrates four information operations to maximize task-relevant signals under finite context budgets: source compression via blueprint distillation, structured indexing using stateful code memory, conditional knowledge injection via retrieval-augmented generation, and closed-loop error correction. Extensive evaluations on the PaperBench benchmark demonstrate that DeepCode achieves state-of-the-art performance, decisively outperforming leading commercial agents such as Cursor and Claude Code, and crucially, surpassing PhD-level human experts from top institutes on key reproduction metrics. By systematically transforming paper specifications into production-grade implementations comparable to human expert quality, this work establishes new foundations for autonomous scientific reproduction that can accelerate research evaluation and discovery.