SWE-Bench Verified
A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.
Claude Mythos Preview from Anthropic currently leads the SWE-Bench Verified leaderboard with a score of 0.939 across 90 evaluated AI models.
Claude Mythos Preview leads with 93.9%, followed by Claude Opus 4.7 at 87.6% and Claude Opus 4.5 at 80.9%.
Progress Over Time
Interactive timeline showing model performance evolution on SWE-Bench Verified
SWE-Bench Verified Leaderboard
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 | Anthropic | — | — | — | ||
| 2 | Anthropic | — | 1.0M | $5.00 / $25.00 | ||
| 3 | Anthropic | — | — | — | ||
| 4 | Anthropic | — | 1.0M | $5.00 / $25.00 | ||
| 5 |  Google | — | 1.0M | $2.50 / $15.00 | ||
| 5 |  DeepSeek | 1.6T | 1.0M | $1.74 / $3.48 | ||
| 7 |  MiniMax | 230B | 1.0M | $0.30 / $1.20 | ||
| 7 |  Moonshot AI | 1.0T | 262K | $0.95 / $4.00 | ||
| 9 |  OpenAI | — | 400K | $1.75 / $14.00 | ||
| 10 | Anthropic | — | 200K | $3.00 / $15.00 | ||
| 11 | DeepSeek | 284B | 1.0M | $0.14 / $0.28 | ||
| 12 |  Alibaba Cloud / Qwen Team | — | 1.0M | $0.50 / $3.00 | ||
| 13 | Google | — | 1.0M | $0.50 / $3.00 | ||
| 13 |  Xiaomi | 1.0T | — | — | ||
| 15 |  Zhipu AI | 744B | 200K | $1.00 / $3.20 | ||
| 16 |  Mistral AI | 128B | 256K | $1.50 / $7.50 | ||
| 17 |  Meta | — | — | — | ||
| 18 | Alibaba Cloud / Qwen Team | 28B | 262K | $0.60 / $3.60 | ||
| 19 | Moonshot AI | 1.0T | — | — | ||
| 20 |  ByteDance | — | — | — | ||
| 21 | Alibaba Cloud / Qwen Team | 397B | 262K | $0.60 / $3.60 | ||
| 22 | OpenAI | — | 400K | $1.25 / $10.00 | ||
| 22 | OpenAI | — | 400K | $1.25 / $10.00 | ||
| 22 | OpenAI | — | — | — | ||
| 25 | Google | — | — | — | ||
| 26 | OpenAI | — | — | — | ||
| 27 | Xiaomi | — | — | — | ||
| 28 | OpenAI | — | — | — | ||
| 28 | Anthropic | — | — | — | ||
| 30 |  StepFun | 196B | 66K | $0.10 / $0.40 | ||
| 31 | Zhipu AI | 358B | — | — | ||
| 32 | OpenAI | — | — | — | ||
| 33 | ByteDance | — | — | — | ||
| 34 | Alibaba Cloud / Qwen Team | 35B | — | — | ||
| 34 | Xiaomi | 309B | — | — | ||
| 36 | Anthropic | — | 200K | $1.00 / $5.00 | ||
| 37 | DeepSeek | 685B | — | — | ||
| 37 | DeepSeek | 685B | — | — | ||
| 37 | DeepSeek | 685B | — | — | ||
| 40 | Anthropic | — | — | — | ||
| 41 | Anthropic | — | — | — | ||
| 42 | Alibaba Cloud / Qwen Team | 27B | 262K | $0.30 / $2.40 | ||
| 43 | Alibaba Cloud / Qwen Team | 122B | 262K | $0.40 / $3.20 | ||
| 44 | Moonshot AI | 1.0T | — | — | ||
| 45 | — | 256K | $0.20 / $1.50 | |||
| 46 | Anthropic | — | — | — | ||
| 47 |  Meituan | 560B | — | — | ||
| 48 | Alibaba Cloud / Qwen Team | 480B | — | — | ||
| 48 | Alibaba Cloud / Qwen Team | 1.0T | — | — | ||
| 50 | MiniMax | 230B | 1.0M | $0.30 / $1.20 |
FAQ
Common questions about SWE-Bench Verified.
Sub-benchmarks
SWE-Bench Multimodal
SWE-Bench Multimodal extends SWE-Bench to evaluate language models on software engineering tasks that involve visual inputs such as screenshots, UI mockups, and diagrams alongside code understanding.
SWE-Bench Pro
SWE-Bench Pro is an advanced version of SWE-Bench that evaluates language models on complex, real-world software engineering tasks requiring extended reasoning and multi-step problem solving.
More evaluations to explore
Related benchmarks in the same category
A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.
A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.
All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.
Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains
Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions
LiveCodeBench is a holistic and contamination-free evaluation benchmark for large language models for code. It continuously collects new problems from programming contests (LeetCode, AtCoder, CodeForces) and evaluates four different scenarios: code generation, self-repair, code execution, and test output prediction. Problems are annotated with release dates to enable evaluation on unseen problems released after a model's training cutoff.