1
Claude Opus 5.5
57.6
—
—
评测基准榜
Artificial Analysis 公开指数 · 分数越高越好 · 点击可进模型详情 · 最后更新 今天 00:48
名次模型综合智能写代码当助手
2
Claude Sonnet 5.5
56
—
—
3
Claude Fable 5.1
53.4
81.6
57.9
4
Qwen3.8 Max
53.4
68.9
49.9
5
GPT-6 Astra
52.7
76.9
51
6
Gemini 4 Argon
52.6
—
—
7
GPT-6.1 Sol
51.8
—
—
8
Claude Opus 5
50.8
78
56.5
9
Claude Fable 5
49.6
76.5
50.7
10
Muse Spark 1.3
48.1
75.8
55.5
11
GPT-6 Sol
47.6
—
—
12
GPT-5.6 Sol
47
77.4
50.2
13
Grok 4.7
46.4
—
—
14
MiMo-V2.6-Pro
46.3
—
—
15
Qwen3.8 Max
45.4
76.2
56
16
GLM-5.3
44.8
74.8
53.1
17
Grok 4.6
44.3
76.8
53
18
Kimi K3
43.6
76.2
50
19
Claude Haiku 5.5
43.4
—
—
20
GPT-5.6 Terra
42.1
76.7
43.2
21
Claude Opus 4.8
41.8
74.3
41.9
22
GLM-5.3-Flash
41.8
71.5
50.9
23
Ling 3.1 Flash
41.1
—
—
24
Gemini 3.8 Flash
40.9
76.3
40.2
25
Claude Opus 4.7
40.7
73.6
38.6
26
Qwen3.8 2.4T A95B
39.9
71.9
50.1
27
Gemini 3.7 Flash
39.6
71.5
—
28
Muse Spark 1.2
39.6
72.2
43.2
29
DeepSeek V4.1 Flash
39.5
—
—
30
GPT-5.4
39
71.1
—
31
Grok 4.5
38.8
72.4
41.2
32
GPT-5.5
38.4
74.9
36.4
33
Claude Sonnet 5
38.2
71.5
43.6
34
GPT-6 Luna
38.1
—
—
35
MiMo-V2.6-Flash
37.9
—
—
36
GPT-5.6 Luna
37.3
71.4
42.1
37
DeepSeek V4 Pro 0813
36
68.8
41.3
38
DeepSeek V4 Flash Vision
34.8
65
47.5
39
DeepSeek V4 Flash 0731
34.3
69.1
41
40
Gemini 3.6 Flash
34
69.2
29
41
Muse Spark 1.1
33.7
71.3
25.8
42
Qwen3.8 27B
33.7
68.1
45.8
43
GLM-5.2
33.7
68.8
38.4
44
Gemini 3.5 Flash
33.6
—
—
45
GPT-5.3 Codex
32.5
—
—
46
Claude Opus 4.6
31.9
—
—
47
DeepSeek V4 Pro 0424
30.4
59.4
26.3
48
GPT-5.2
30.4
—
—
49
Claude Sonnet 4.6
30.1
63
31.8
50
Gemini 3.1 Pro Preview
29.7
68.8
8.2
51
Qwen3.7 Max
29.5
66
22.5
52
MiniMax-M3
29.2
58.6
29.5
53
Claude Opus 4.5
29.1
—
—
54
MiMo-V2-Pro
28.6
—
—
55
GPT-5.2 Codex
28.5
—
—
56
Qwen3.6 Max Preview
28.4
—
—
57
Nex-N2-Pro
28.2
59.1
23.7
58
Solar Pro 4
28.2
52.7
—
59
GLM-5
27.9
—
—
60
Grok Build 0.1 0616
27.2
51.5
—
61
Kimi K2.6
27
61.8
20.5
62
Qwen3.6 Plus
27
54.5
—
63
GLM-5-Turbo
26.6
—
—
64
Gemini 3 Flash Preview
26.3
—
—
65
GLM-5.1
26.1
55.8
23.9
66
MiMo-V2.5-Pro
26
60.2
21.3
67
Kimi K2.7 Code
25.8
60.8
21
68
Inkling Small
25.7
52.9
23.5
69
Grok 4.20 0309 v2
25.7
—
—
70
Hy3
25.3
58.8
24.1
71
Qwen3.7 Plus
25.2
55.9
17.5
72
MiMo-V2.5
25.2
56.8
15.8
73
MiMo-V2-Omni-0327
25.1
—
—
74
Inkling
25
52.1
22.5
75
GPT-5 Codex
24.9
—
—
76
Grok 4.3
24.9
42.2
15.5
77
GPT-5.1
24.7
49.4
—
78
Ling-3.0-flash-VL
24.6
57
28.7
79
DeepSeek V4 Flash 0420
24.4
52
26.3
80
GPT-5.4 mini
24.1
56.1
17.9
81
Solar Mini 4
24.1
—
—
82
GPT-5.1 Codex
23.7
—
—
83
Kimi K2.5
23.5
46.8
—
84
GLM 5V Turbo
23.5
—
—
85
GPT-5
23
37.8
—
86
Nemotron 3 Ultra 550B A55B
22.9
49.3
20.1
87
Qwen3.5 27B
22.9
—
—
88
Claude 4.1 Opus
22.8
—
—
89
MiniMax-M2.5
22.8
—
—
90
MiniMax-M2.7
22.8
52.6
15.3
91
Ling-3.0-flash-Fin
22.6
55.6
27.9
92
Grok 4
22.5
—
—
93
MiMo-V2-Flash
22.4
—
—
94
Gemini 3.5 Flash-Lite
22.2
49.3
14.3
95
GLM-4.7
22.2
45.3
—
96
Kimi K2 Thinking
22
—
—
97
o3-pro
21.9
—
—
98
KAT Coder Pro V2
21.7
59.5
—
99
DeepSeek V3.2
21.5
44.2
—
100
Qwen3.5 397B A17B
21.4
—
—
AI++指数
全部前沿
全部最新模型
行业动态
全部- 产业观察 · 2026-09-13用量榜上的「可控 AI 编程」：CodeGPT 类多模型 + 私有部署编程助手，正在改写 IDE 插件之争
- 产业观察 · 2026-09-09个人 Agent 网关成新品类：OpenClaw 养虾热潮与自托管智能体的入口之争
- 竞争格局 · 2026-09-08开源 Coding Agent 2026：Qwen Code 入场，Claude Code 的护城河被三条路同时撬动
- 产业观察 · 2026-09-01AI 3D 生成迎来产业化分水岭：VAST 30 亿融资 + Tripo P2.0，影眸同日发布世界生成模型
- 产业观察 · 2026-09-05Bun 1.4 把运行时变成『全栈平台』：JS 运行时之战从『跑得快』卷到『什么都不用装』