Coding Evaluation Benchmark

SWE-bench Verified

SWE-bench: Resolving Real-World GitHub Issues (Verified Subset)

Standard benchmark evaluating an AI model's ability to solve end-to-end software engineering problems from real GitHub issues in popular Python repositories.

Verified Model Leaderboard

Ranked by verified Pass@1 (%) score from official technical reports.

Primary Metric: Pass@1 (%)
RankAI Model & ProviderScore (Pass@1 (%))Evaluation MethodologyVerification DateSource
#1Claude 3.7 Sonnet

Anthropic Hybrid Reasoning Transformer

70.3%Pass@1, Thinking mode enabled with Claude Code scaffolding and 64k token thinking budgetFeb 2025Anthropic Claude 3.7 Sonnet Announcement
#2OpenAI o3-mini

OpenAI Reinforcement Learning Reasoning Model

49.3%Pass@1 with high reasoning effortJan 2025OpenAI o3-mini Announcement
#3DeepSeek-R1

DeepSeek MoE (Mixture of Experts) Reasoning

49.2%Pass@1 with standard SWE-bench frameworkJan 2025DeepSeek-R1 Technical Report
#4Claude 3.5 Sonnet (v2)

Anthropic Dense Transformer

49.0%Pass@1, standard agentic scaffolding without reasoning tokensOct 2024Anthropic Technical Announcement
#5OpenAI o1

OpenAI Reinforcement Learning Reasoning Model

48.9%Pass@1, high reasoning effort without external agent scaffoldingDec 2024OpenAI o1 System Card
#6Claude 3.5 Haiku

Anthropic Dense Transformer

40.6%Pass@1 with standard developer scaffoldNov 2024Anthropic Announcement
#7GPT-4o

OpenAI Multimodal Transformer

38.8%Pass@1 standard harnessAug 2024SWE-bench Official Leaderboard
#8Qwen 2.5 Coder 32B Instruct

Qwen (Alibaba Cloud) Dense Transformer

35.8%Pass@1 with standard open-weights harnessNov 2024Qwen 2.5 Coder Technical Report
#9GPT-4o mini

OpenAI Multimodal Small Transformer

20.2%Pass@1 standard scaffoldJul 2024SWE-bench Leaderboard

Evaluation Methodology & Harness

Verified subset filters out 500 hand-checked, unambiguous problem instances with comprehensive unit tests executed in sandboxed Docker containers.

Dataset & Test Set Construction

500 verified real-world GitHub issues across popular open-source repositories including Django, SymPy, scikit-learn, and pytest.

Why This Benchmark Matters

The gold standard for evaluating real-world coding agents, requiring multi-file codebase navigation, bug localization, and patch generation.

Known Limitations & Contamination Risks

Focused primarily on Python; execution speed and test-harness timeouts can affect reproducibility without standardized scaffolding.

Data Accuracy & Verification Notice

AI model specifications, pricing records, and benchmark metrics published on this platform are compiled directly from authoritative sources (official provider documentation, research papers, and verified evaluation harnesses). Benchmark results reflect specific test harnesses and prompting methodologies; scores are not directly comparable across differing evaluation setups.