AI EVALUATION, IN CONTEXT.BY EQUALITY MACHINE
Benchmark Radar.Benchmarks. Model results. Original sources.
/

LLM Stats

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.

评测信息与证据

来源
LLM Stats
发布者
Scale AI
评测发布日期
2025-09-05
采集日期
2026-08-17
模态
text

本站保留来源报告的结果，不代表独立复测。请先核对任务版本、评测协议、分数单位和报告日期；不同运行条件不能直接合并为统一排名。

原始资料

报告成绩 50

按来源中的观察顺序展示，顺序不代表排名。日期保留原始精度。

模型数值单位协议报告日期与口径来源
Claude Fable 50.8未收录未收录2026-06-09 · model_announcementLLM Stats
Claude Mythos Preview0.778未收录未收录2026-04-07 · model_announcementLLM Stats
Claude Opus 4.80.692未收录未收录2026-05-28 · model_announcementLLM Stats
Qwen3.8 Max0.677未收录未收录2026-08-02 · model_announcementLLM Stats
Grok 4.50.647未收录未收录2026-07-16 · model_announcementLLM Stats
GPT-5.6 Sol0.646未收录未收录2026-07-09 · model_announcementLLM Stats
Claude Opus 4.70.643未收录未收录2026-04-16 · model_announcementLLM Stats
GPT-5.6 Terra0.634未收录未收录2026-07-09 · model_announcementLLM Stats
Claude Sonnet 50.632未收录未收录2026-06-30 · model_announcementLLM Stats
GPT-5.6 Luna0.627未收录未收录2026-07-09 · model_announcementLLM Stats
GLM-5.20.621未收录未收录2026-06-16 · model_announcementLLM Stats
Qwen3.8-27B0.617未收录未收录2026-08-14 · model_announcementLLM Stats
Muse Spark 1.10.615未收录未收录2026-07-09 · model_announcementLLM Stats
Qwen3.7 Max0.606未收录未收录2026-05-19 · model_announcementLLM Stats
Laguna S 2.10.594未收录未收录2026-07-21 · model_announcementLLM Stats
MiniMax M30.59未收录未收录2026-06-01 · model_announcementLLM Stats
Gemini 3.6 Flash0.587未收录未收录2026-07-21 · model_announcementLLM Stats
GPT-5.50.586未收录未收录2026-04-23 · model_announcementLLM Stats
Kimi K2.60.586未收录未收录2026-04-20 · model_announcementLLM Stats
GLM-5.10.584未收录未收录2026-04-07 · model_announcementLLM Stats
Hy30.579未收录未收录2026-07-06 · model_announcementLLM Stats
GPT-5.40.577未收录未收录2026-03-05 · model_announcementLLM Stats
Qwen3.7-Plus0.576未收录未收录2026-05-31 · model_announcementLLM Stats
Seed 2.1 Pro0.575未收录未收录2026-06-24 · model_announcementLLM Stats
MiMo-V2.5-Pro0.572未收录未收录2026-04-27 · model_announcementLLM Stats
Seed 2.1 Turbo0.57未收录未收录2026-06-24 · model_announcementLLM Stats
GPT-5.3 Codex0.568未收录未收录2026-02-05 · model_announcementLLM Stats
Qwen3.6 Plus0.566未收录未收录2026-04-02 · model_announcementLLM Stats
GPT-5.2 Codex0.564未收录未收录2026-01-14 · model_announcementLLM Stats
MiniMax M2.70.562未收录未收录2026-03-18 · model_announcementLLM Stats
MiMo-V2.50.561未收录未收录2026-04-22 · model_announcementLLM Stats
Inkling-Small0.559未收录未收录2026-07-30 · model_announcementLLM Stats
DeepSeek-V4-Pro-Max0.554未收录未收录2026-04-23 · model_announcementLLM Stats
MiniMax M2.50.554未收录未收录2026-02-12 · model_announcementLLM Stats
Gemini 3.5 Flash0.551未收录未收录2026-05-19 · model_announcementLLM Stats
GPT-5.4 mini0.544未收录未收录2026-03-17 · model_announcementLLM Stats
Gemini 3.5 Flash-Lite0.542未收录未收录2026-07-21 · model_announcementLLM Stats
Gemini 3.1 Pro0.542未收录未收录2026-02-19 · model_announcementLLM Stats
Qwen3.6-27B0.535未收录未收录2026-04-21 · model_announcementLLM Stats
MAI-Thinking-10.528未收录未收录2026-06-02 · model_announcementLLM Stats
DeepSeek-V4-Flash-Max0.526未收录未收录2026-04-23 · model_announcementLLM Stats
Muse Spark0.524未收录未收录2026-04-08 · model_announcementLLM Stats
GPT-5.4 nano0.524未收录未收录2026-03-17 · model_announcementLLM Stats
DeepSeek-V4-Flash-04230.523未收录未收录2026-04-23 · model_announcementLLM Stats
Muse Glimmer-30B0.512未收录未收录2026-08-10 · model_announcementLLM Stats
MAI-Code-1-Flash0.512未收录未收录2026-06-02 · model_announcementLLM Stats
Kimi K2.50.507未收录未收录2026-01-27 · model_announcementLLM Stats
Qwen3.6-35B-A3B0.495未收录未收录2026-04-16 · model_announcementLLM Stats
Laguna XS 2.10.476未收录未收录2026-07-02 · model_announcementLLM Stats
North Mini Code 1.00.402未收录未收录2026-06-09 · model_announcementLLM Stats

继续阅读

编程评测工具与智能体评测推理与数学评测