DataLearner logoDataLearnerAI
Latest AI Insights
Model Leaderboards
Benchmarks
Model Directory
Model Comparison
Resource Center
Tools
LanguageEnglish
DataLearner logoDataLearner AI

A knowledge platform focused on LLM benchmarking, datasets, and practical instruction with continuously updated capability maps.

Products

  • Leaderboards
  • Model comparison
  • Datasets

Resources

  • Tutorials
  • Editorial
  • Tool directory

Company

  • About
  • Privacy policy
  • Data methodology
  • Contact

© 2026 DataLearner AI. DataLearner curates industry data and case studies so researchers, enterprises, and developers can rely on trustworthy intelligence.

Privacy policyTerms of service
  1. Back to benchmark list
  2. /
  3. SWE-Bench Pro - Public

SWE-Bench Pro - Public

Scale AI 于 2025 年 9 月 21 日发布了 SWE-Bench Pro,这是一个针对 AI 代理在软件工程任务上的评估基准。该基准包含 1,865 个问题,来源于 41 个活跃维护的代码仓库,聚焦企业级复杂任务。现有模型在该基准上的表现显示出显著差距,顶级模型的通过率低于 25%,而最近的榜单更新显示部分模型已超过 40%。这一发布旨在推动 AI 在长时程软件开发中的应用研究。

更新于 2026-04-22
5,858 次浏览
问题数量
731
发布机构
Scale AI
评测类别
编程与软件工程
评测指标
Accuracy
支持语言
英文
难度等级
高难度

简介

一个用于评估大模型解决真实世界代码工程的评测基准

相关资源

查看原始论文
阅读学术论文原文
获取数据集
下载评测数据集
访问官网
浏览项目官方网站
DataLearner 介绍
中文详细解读

SWE-Bench Pro - Public Model Score Leaderboard

Source: DataLearnerAI

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

模式说明:
normal
thinking
low
medium
high
deeper thinking
parallel_thinking
图表加载中...

Latest SWE-Bench Pro - Public model rankings and full benchmark leaderboard

Browse the latest scores, model modes, release dates, and parameter sizes for SWE-Bench Pro - Public.

Model release cutoff:

SWE-Bench Pro - Public详细排名数据表格

排名模型
1
Claude Mythos Preview
Extended ThinkingTools
77.802026-04-07未知
2
Opus 4.7
Extended ThinkingTools
64.302026-04-16未知
3
Kimi K2.6
Thinking EnabledTools
58.602026-04-2010000
4
GLM 5.1
Thinking EnabledTools
58.402026-03-27754
5
GPT-5.4
Thinking Level · Extra High
57.702026-03-05未知
6
Qwen3.6-Max-Preview
Deep Thinking ModeTools
57.302026-04-2010000
7
GPT-5.3 Codex
Thinking Enabled
56.802026-02-05未知
8
Qwen 3.6 Plus Preview
Thinking EnabledTools
56.602026-03-31未知
9
Qwen 3.6 Plus Preview
Thinking Enabled
56.602026-03-31未知
10
MiniMax-M2.7
Thinking EnabledTools
56.202026-03-182290
11
GPT-5.2
Thinking Level · Extra HighTools
55.602025-12-11未知
12
MiniMax M2.5
Thinking EnabledTools
55.402026-02-122290
13
GPT-5.4 mini
Thinking Level · Extra HighTools
54.402026-03-17未知
14
Gemini 3.1 Pro Preview
Thinking Level · HighTools
54.202026-02-20未知
15
Qwen3.6-27B
Thinking EnabledTools
53.502026-04-22270
16
GPT-5.4 nano
Thinking Level · Extra HighTools
52.402026-03-17未知
17
Qwen3.5-397B-A17B
Thinking Enabled
50.902026-02-16397
18
GPT-5.1
Thinking Level · High
50.802025-11-12未知
19
Kimi K2.5
Thinking EnabledTools
50.702026-01-2710000
20
Qwen3.6-35B-A3B
Thinking Enabled
49.502026-04-16350
21
Qwen3-Coder-Next
Standard ModeTools
44.302026-02-0380
22
Claude Sonnet 4.5
Thinking Enabled
43.602025-09-30未知
23
Claude Sonnet 4
Thinking Enabled
42.702025-05-23未知
24
DeepSeek V3.2
Thinking Enabled
40.902025-12-016710
25
GLM-4.7
Thinking EnabledTools
40.602025-12-223580
26
Haiku 4.5
Extended ThinkingTools
39.452025-10-15未知
27
GPT-5
Thinking Level · High
36.302025-08-07未知
28
M2.1
Thinking EnabledTools
32.602025-12-232300
29
Kimi K2 0905
Standard Mode
27.672025-09-0510000