# SWE-fficiency: Can Language Models Optimize Real-World Repositories on Real Workloads?

arXiv: https://arxiv.org/abs/2511.06090 · HTML: https://mlsanity.com/paper/2511.06090/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 37 references that are in the map.
- First posted on arXiv 8 Nov 2025.

## Builds on (most cited first, 17 of 17)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [Training Software Engineering Agents and Verifiers with SWE-Gym](https://mlsanity.com/paper/2412.21139/) (2024)
- [KernelBench: Can LLMs Write Efficient GPU Kernels?](https://mlsanity.com/paper/2502.10517/) (2025)
- [SWE-bench Multimodal: Do AI Systems Generalize to Visual Software Domains?](https://mlsanity.com/paper/2410.03859/) (2024)
- [SWT-Bench: Testing and Validating Real-World Bug-Fixes with Code Agents](https://mlsanity.com/paper/2406.12952/) (2024)
- [Learning Performance-Improving Code Edits](https://mlsanity.com/paper/2302.07867/) (2023)
- [Evaluating Language Models for Efficient Code Generation](https://mlsanity.com/paper/2408.06450/) (2024)
- [EffiBench: Benchmarking the Efficiency of Automatically Generated Code](https://mlsanity.com/paper/2402.02037/) (2024)
- [SWE-Perf: Can Language Models Optimize Code Performance on Real-World Repositories?](https://mlsanity.com/paper/2507.12415/) (2025)
- [Mercury: A Code Efficiency Benchmark for Code Large Language Models](https://mlsanity.com/paper/2402.07844/) (2024)
- [ECCO: Can We Improve Model-Generated Code Efficiency Without Sacrificing Functional Correctness?](https://mlsanity.com/paper/2407.14044/) (2024)
- [AlgoTune: Can Language Models Speed Up General-Purpose Numerical Programs?](https://mlsanity.com/paper/2507.15887/) (2025)
- [GSO: Challenging Software Optimization Tasks for Evaluating SWE-Agents](https://mlsanity.com/paper/2505.23671/) (2025)

## Built on by (most cited first, 21 of 21)

- [Immersion in the GitHub Universe: Scaling Coding Agents to Mastery](https://mlsanity.com/paper/2602.09892/) (2026)
- [BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing?](https://mlsanity.com/paper/2603.03194/) (2026)
- [Confucius Code Agent: Scalable Agent Scaffolding for Real-World Codebases](https://mlsanity.com/paper/2512.10398/) (2025)
- [Agentic Software Issue Resolution with Large Language Models: A Survey](https://mlsanity.com/paper/2512.22256/) (2025)
- [FormulaCode: Evaluating Agentic Optimization on Large Codebases](https://mlsanity.com/paper/2603.16011/) (2026)
- [From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI](https://mlsanity.com/paper/2606.14502/) (2026)
- [VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?](https://mlsanity.com/paper/2605.06068/) (2026)
- [PerfAgent: Profiler-Guided Iterative Refinement for Repository-Level Code Optimization](https://mlsanity.com/paper/2607.19653/) (2026)
- [Extrapolative Weight Averaging Reveals Correctness-Efficiency Frontiers in Code RL](https://mlsanity.com/paper/2605.28751/) (2026)
- [Themis: Training Robust Multilingual Code Reward Models for Flexible Multi-Criteria Scoring](https://mlsanity.com/paper/2605.00754/) (2026)
- [Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?](https://mlsanity.com/paper/2607.01211/) (2026)
- [SecProbe: Adaptive Evaluation of Coding Agents on Cybersecurity Vulnerabilities](https://mlsanity.com/paper/2609.33763/) (2026)
- [SemRep: Generative Code Representation Learning with Code Transformations](https://mlsanity.com/paper/2603.13640/) (2026)
- [Post-training is (Massive) Supervised Learning](https://mlsanity.com/paper/2606.07527/) (2026)
- [PERFOPT-Bench: Evaluating Coding Agents on Software Performance Optimization](https://mlsanity.com/paper/2607.07744/) (2026)
- [Reinforcement Learning for Code Optimization](https://mlsanity.com/paper/2607.25970/) (2026)
- [Is manual software optimization a thing of the past?](https://mlsanity.com/paper/2609.37849/) (2026)
- [Correct Code, Broken Contributions? SWE-CC: Benchmarking Repository Policy Compliance for Coding Agents](https://mlsanity.com/paper/2610.06193/) (2026)
- [SWE-Journey: Towards More Realistic Evaluation of Coding Assistants through Long-Horizon, Multi-Turn Interaction](https://mlsanity.com/paper/2610.11559/) (2026)
- [Not All Correct Fixes Run Alike: An Empirical Study of Patch Performance in Repository-Level Issue Resolution](https://mlsanity.com/paper/2610.13563/) (2026)
- [DuoDiscover: Scaling Autonomous Discovery Beyond the Evaluation Bottleneck](https://mlsanity.com/paper/2610.17184/) (2026)

## Abstract (arXiv)

Optimizing the performance of large-scale software repositories demands expertise in code reasoning and software engineering (SWE) to reduce runtime while preserving program correctness. However, most benchmarks emphasize what to fix rather than how to fix code. We introduce SWE-fficiency, a benchmark for evaluating repository-level performance optimization on real workloads. Our suite contains 498 tasks across nine widely used data-science, machine-learning, and HPC repositories (e.g., numpy, pandas, scipy): given a complete codebase and a slow workload, an agent must investigate code semantics, localize bottlenecks and relevant tests, and produce a patch that matches or exceeds expert speedup while passing the same unit tests. To enable this how-to-fix evaluation, our automated pipeline scrapes GitHub pull requests for performance-improving edits, combining keyword filtering, static analysis, coverage tooling, and execution validation to both confirm expert speedup baselines and identify relevant repository unit tests. Empirical evaluation of state-of-the-art agents reveals significant underperformance. On average, agents achieve less than 0.23x the expert speedup: agents struggle in localizing optimization opportunities, reasoning about execution across functions, and maintaining correctness in proposed edits. We release the benchmark and accompanying data pipeline to facilitate research on automated performance engineering and long-horizon software reasoning.