跳到主内容
知行录
leaderboard.cn
Toggle Sidebar
搜索
⌘K
分析
前沿模型
历史趋势
本地模型
资料库
模型库
厂商
基准评测
数据源
显示模式
Toggle Sidebar
知行录
leaderboard.cn
正在加载页面内容
正在加载页面内容
首页
Agent
面向工具调用、网页浏览和长流程执行。
更新于 07/16 02:55
模型
118 个 · 16 项基准
评分
BrowseComp
Arena Agent Bash Recovery
Arena Agent Confirmed Success
Arena Agent Net Improvement
Arena Agent Praise vs Complaint
Arena Agent Steerability
Arena Agent Tool Hallucination
Terminal-Bench 2.1
Toolathlon
CyberGym
HLE (w/ Tools)
Artificial Analysis Intelligence Index v4.1
AutomationBench
GDPval-AA v2
Agents' Last Exam
Management Consulting Tasks (Internal)
1
GPT-5.6 Sol
100.0
90.4%
7.5%
10.9%
10.9%
17.6%
17.3%
1.3%
88.8%
58%
84.5%
-
58.9 pts
18.1%
1747.8 pts
52.7%
43.2%
2
Claude Fable 5
98.8
88%
9.4%
17.3%
13.9%
30.6%
12.1%
1.3%
84.3%
61.7%
83.8%
-
59.9 pts
17.4%
1759.6 pts
40.5%
35.5%
3
Claude Opus 4.8
93.9
84.3%
9.9%
8.6%
9.4%
18.0%
10.3%
0.6%
82.7%
59.9%
78.1%
57.9%
55.7 pts
15.5%
1600.1 pts
45.2%
31.6%
4
GPT-5.5
93.5
84.4%
14.5%
6.9%
8.3%
13.6%
6.7%
1.3%
85.6%
55.6%
81.8%
52.2%
54.8 pts
12.9%
1493.7 pts
46.9%
31.3%
5
GPT-5.6 Terra
93.2
87.5%
-
-
-
-
-
-
87.4%
53.1%
81.8%
-
55 pts
15.2%
1593 pts
50.4%
37.2%
6
GPT-5.5 Pro
92.5
90.1%
-
-
-
-
-
-
-
-
-
57.2%
-
-
-
-
-
7
GPT-5.4 Pro
91.6
89.3%
-
-
-
-
-
-
-
-
-
58.7%
-
-
-
-
-
8
Claude Opus 4.7
90.8
79.3%
11.7%
5.5%
8.2%
13.7%
9.2%
1.3%
78.9%
-
73.1%
54.7%
-
-
-
-
-
9
Grok 4.5
90.2
-
9.7%
6.9%
5.0%
6.1%
0.9%
1.3%
83.3%
-
-
-
-
-
-
-
-
10
Claude Sonnet 5
89.5
-
9.6%
10.3%
8.0%
13.9%
4.9%
1.2%
-
-
-
-
-
-
-
-
-
11
GLM-5.2
89.1
-
5.2%
9.1%
6.9%
15.4%
4.1%
1.3%
-
48.2%
-
54.7%
-
-
-
-
-
12
Seed2.1 Pro
88.1
86.2%
-
-
-
-
-
-
71%
50.6%
-
55.7%
-
-
-
-
-
13
Claude Opus 4.6
87.9
84%
11.8%
3.5%
6.5%
9.4%
6.7%
1.3%
-
47.2%
66.6%
53.1%
-
-
-
-
-
14
GPT-5.4
87.2
82.7%
11.3%
6.6%
6.7%
6.1%
8.0%
1.3%
-
54.6%
79%
52.1%
-
-
-
-
-
15
GPT-5.6 Luna
86.9
83.3%
-
-
-
-
-
-
84.7%
53.4%
77.9%
-
51.2 pts
14.9%
1591.8 pts
50.3%
35.4%
16
混元3
86.8
84.2%
-
-
-
-
-
-
-
48.5%
-
53.2%
-
-
-
-
-
17
Muse Spark 1.1
83.9
-
4.0%
2.1%
0.2%
0.6%
-7.2%
1.3%
80%
-
-
-
-
-
-
-
-
18
Kimi-K2.7-Code
83.5
-
-0.2%
4.5%
-0.8%
1.4%
-1.6%
1.3%
-
-
-
-
-
-
-
-
-
19
MiMo-V2.5
81.9
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
20
Claude Sonnet 4.6
81.7
74.7%
11.6%
-0.9%
2.2%
-0.8%
1.3%
1.3%
-
-
-
-
-
-
-
-
-
21
Kimi-K2.6
81.1
83.2%
-1.8%
2.4%
-1.8%
-0.2%
-4.1%
1.3%
-
50%
-
54%
-
-
-
-
-
22
LongCat-2.0
81.0
79.9%
-
-
-
-
-
-
70.8%
-
-
-
-
-
-
-
-
23
MiniMax M3
81.0
83.5%
6.3%
-2.1%
-2.3%
-7.2%
-5.1%
1.3%
-
-
-
-
-
-
-
-
-
24
Seed2.1 Turbo
80.9
-
-
-
-
-
-
-
67.6%
-
-
-
-
-
-
-
-
低
高
单一有效来源基准按 0.5 权重计入,Arena.ai 除外;最高模型归一化为 100