跳过主要内容
评测来源/Terminal-Bench 4.0(AA 复跑)
参与排名编程

Terminal-Bench 4.0(AA 复跑)

Artificial Analysis · Terminal-Bench 4.0(AA 复跑) 的公开模型评测结果。

官方评测
来源说明

该来源测量模型在「编程」领域的具体交付表现,原始成绩会和统一能力读数分开展示。

限制: 不同版本、推理设置和测试集可能导致分数变化,请结合数据时间阅读。

使用方式

参与排名的来源会影响能力分;参考、交叉参考和观察来源仅用于对比与核验。

本来源的模型成绩

Claude Sonnet 5.50.6363636489.12026-11-05
GPT-6 Astra0.595959686.02026-11-05
Claude Opus 5.50.595959686.02026-11-05
Gemini 4 Argon0.5707070784.02026-11-05
Muse Wonder Full0.5707070784.02026-10-17
Muse Wonder Full0.5707070784.02026-10-18
Muse Wonder Full0.5707070784.02026-11-05
GPT-6.1 Sol0.5606060683.22026-11-05
GPT-6.1 Terra0.5555555682.82026-11-05
Claude Fable 5.10.5505050582.42026-11-05
Muse Wonder0.5202020280.02026-10-17
Muse Wonder0.5202020280.02026-11-05
Muse Wonder0.5202020280.02026-10-18
Claude Opus 50.4898989977.62026-11-05
GPT-6 Sol0.4393939473.62026-11-05
Claude Fable 50.4242424272.42026-11-05
GLM 5.30.4191919272.02026-11-05
GPT-5.6 Sol0.398989970.42026-11-05
GPT-5.6 Terra0.3535353566.82026-11-05
MiMo V2.6 Pro0.3484848566.42026-11-05
DeepSeek V4.1 Flash 10170.3383838465.62026-11-05
Ling-3.1-flash0.3333333365.22026-10-18
Ling-3.1-flash0.3333333365.22026-11-05
Ling-3.1-flash0.3333333365.22026-10-17
Muse Spark 1.30.3333333365.22026-11-05
Step 5 Preview0.3333333365.22026-11-05
Ling-3.1-flash-10190.3282828364.82026-11-05
GLM-5.3 Flash0.3282828364.82026-10-17
GLM-5.3 Flash0.3282828364.82026-11-05
GLM-5.3 Flash0.3282828364.82026-10-18
Claude Haiku 5.50.3282828364.82026-11-05
DeepSeek V4.1 Flash0.2676767760.02026-11-05
Mistral Large 40.2676767760.02026-10-28
Grok 4.70.2575757659.22026-11-05
MiniMax M3.1 Flash0.2575757659.22026-11-05
Mistral Large 40.2474747558.42026-11-05
MiMo V2.6 Flash0.2272727356.82026-10-17
MiMo V2.6 Flash0.2272727356.82026-11-05
MiMo V2.6 Flash0.2272727356.82026-10-18
Claude Opus 4.80.2171717256.12026-11-05
Grok 4.60.2121212155.72026-11-05
Gemini 3.8 Flash0.196969754.52026-11-05
Qwen3.8 Max0.1868686953.72026-11-05
GPT-5.50.1464646550.52026-11-05
DeepSeek V4 Pro0.1464646550.52026-11-05
Claude Sonnet 50.1414141450.12026-11-05
DeepSeek V4 Pro 08130.1414141450.12026-11-05
Kimi K30.1262626348.92026-11-05
Kimi K30.1262626348.92026-10-27
GPT-6 Luna0.1262626348.92026-11-05
DeepSeek V4 Flash 07310.1212121248.52026-11-05
GPT-5.6 Luna0.1161616248.12026-11-05
Grok 4.50.1060606147.32026-11-05
Gemini 3.6 Flash0.0707070744.52026-11-05
Muse Spark 1.20.0707070744.52026-11-05
Muse Spark 1.10.0606060643.72026-11-05
Gemini 3.1 Pro Preview0.0404040442.12026-11-05
Claude Sonnet 4.60.0303030341.32026-11-05
GPT-5.4 mini0.0202020240.52026-11-05
GLM-5.10.0202020240.52026-11-05
MiniMax-M30.0202020240.52026-11-05
Qwen3.7 Max0.0151515240.12026-11-05
Qwen3.7 Plus0.0101010139.72026-11-05
Kimi K2.7 Code0.0101010139.72026-11-05
GLM-5.20.0101010139.72026-11-05
Inkling0.0101010139.72026-11-05
Gemini 3.5 Flash Lite0.0101010139.72026-11-05
Inkling Small0.0101010139.72026-11-05
Qwen3.5 9B0.0050505139.32026-11-05
Trinity Large Thinking0.0050505139.32026-11-05
Kimi K2.60.0050505139.32026-11-05
Gemini 3.1 Flash Lite0.0050505139.32026-11-05
Ring-2.6-1T0.0050505139.32026-11-05
Command A Plus0.0050505139.32026-11-05
Nemotron 3 Ultra 550B A55B0.0050505139.32026-11-05
North Mini Code0.0050505139.32026-11-05
Hy30.0050505139.32026-11-05
o3-mini038.92026-10-11
DeepSeek-R1038.92026-11-05
o3 Mini High038.92026-11-05
DeepSeek V3 0324038.92026-11-05
Qwen3 32B038.92026-11-05
Llama 4 Maverick038.92026-11-05
Llama 4 Scout038.92026-11-05
Qwen3 8B038.92026-11-05
Qwen3 14B038.92026-11-05
Gemini 2.5 Pro038.92026-11-05
gpt-oss-120b038.92026-11-05
gpt-oss-20b038.92026-11-05
GPT-5 mini038.92026-11-05
DeepSeek V3.1 Terminus038.92026-11-05
Claude Sonnet 4.5038.92026-11-05
Claude Haiku 4.5038.92026-11-05
Mistral Large 3038.92026-11-05
Ministral 3 14B038.92026-11-05
Ministral 3 8B038.92026-11-05
Ministral 3 3B038.92026-11-05
Nemotron 3 Nano 30B A3B038.92026-11-05
Solar Pro 3038.92026-11-05
Qwen3 Coder Next038.92026-11-05
Qwen3.5 397B-A17B038.92026-11-05
Qwen3.5 122B-A10B038.92026-11-05
Mercury 2038.92026-11-05
Nemotron 3 Super 120B A12B038.92026-11-05
MiniMax-M2.7038.92026-11-05
Qwen3.6 35B-A3B038.92026-11-05
Grok 4.3038.92026-11-05
Grok 4.3038.92026-10-27
Qwen3.6 27B038.92026-11-05
MiMo-V2.5038.92026-11-05
MiMo-V2.5-Pro038.92026-11-05
Mistral Medium 3.5038.92026-11-05
LongCat-2.0038.92026-11-05
Ling-3.0-flash (free)038.92026-11-05

常见问题