llm.ing

AA Intelligence Index

run by Artificial Analysisunit: indexDisplay with attribution; no redistribution105 modelssource ↗methodology ↗

Composite of independently run evals by Artificial Analysis.

AA Intelligence IndexAA Coding IndexAA Agentic IndexMedian output speedMedian time to first token
#ModelScoreObserved
1Claude Opus 4.7
55.0
mirror
Aug 7, 2026
2Claude Fable 5.1
53.4
independent
Sep 9, 2026
3GPT-5.4
53.1
mirror
Aug 7, 2026
4GPT-6 Astra
52.7
independent
Sep 20, 2026
5Claude Opus 5
50.8
mirror
Sep 20, 2026
6Claude Fable 5
49.6
mirror
Sep 20, 2026
7GPT-5.6 Sol
47.0
mirror
Sep 20, 2026
8Qwen3.8 Max
45.4
mirror
Sep 16, 2026
9GLM-5.3
44.8
mirror
Sep 20, 2026
10Grok 4.6
44.3
mirror
Sep 20, 2026
11Kimi K3
43.6
mirror
Sep 20, 2026
12GPT-5.6 Terra
42.1
mirror
Sep 20, 2026
13Claude Opus 4.8
41.8
mirror
Sep 20, 2026
13GLM-5.3-Flash
41.8
independent
Sep 20, 2026
15Gemini 3.8 Flash
40.9
mirror
Sep 20, 2026
16Grok Build 0.1
40.7
mirror
Aug 7, 2026
17Qwen3.6 Plus
40.5
mirror
Aug 7, 2026
18Muse Spark 1.2
39.6
independent
Sep 20, 2026
19Gemini 3.7 Flash
39.1
mirror
Sep 20, 2026
20MiniMax-M2.7
38.9
independent
Aug 7, 2026
21Grok 4.5
38.8
mirror
Sep 20, 2026
22GPT-5.5
38.4
independent
Sep 20, 2026
23Claude Sonnet 5
38.2
independent
Sep 20, 2026
24GPT-5.1
37.5
independent
Aug 7, 2026
25GPT-5.6 Luna
37.3
independent
Sep 20, 2026
26DeepSeek V4 Pro
36.0
mirror
Sep 20, 2026
26Kimi K2.5
36.0
mirror
Aug 7, 2026
28GPT-5
35.3
mirror
Aug 7, 2026
29GLM-4.7
34.5
mirror
Aug 7, 2026
30DeepSeek V4 Flash
34.3
mirror
Sep 20, 2026
31Gemini 3.6 Flash
34.0
mirror
Sep 20, 2026
32Muse Spark 1.1
33.7
mirror
Sep 20, 2026
32GLM-5.2
33.7
mirror
Sep 20, 2026
34DeepSeek V3.2
32.6
mirror
Aug 7, 2026
34Gemini 3.5 Flash
32.6
independent
Sep 20, 2026
36Step 3.7 Flash
30.9
mirror
Aug 7, 2026
37GPT-5.2
30.4
independent
Sep 9, 2026
38Claude Sonnet 4.6
30.1
mirror
Sep 20, 2026
39Qwen3.5 27Bnon-reasoning
30.0
independent
Aug 7, 2026
40Gemini 3.1 Pro Preview
29.7
mirror
Sep 20, 2026
41Qwen3.7 Max
29.5
mirror
Sep 20, 2026
42GLM-4.6
29.3
mirror
Aug 7, 2026
43MiniMax-M3
29.2
independent
Sep 20, 2026
44Claude Opus 4.5 (latest)thinking
29.1
independent
Sep 9, 2026
45GPT-5.2 Codex
28.5
independent
Sep 9, 2026
46GLM-5
27.9
independent
Sep 9, 2026
47Kimi K2.6
27.0
mirror
Sep 20, 2026
48Claude Opus 4.6
26.4
independent
Sep 9, 2026
49Gemma 4 26B A4B IT
26.1
mirror
Aug 7, 2026
49GLM-5.1
26.1
mirror
Sep 20, 2026
51Kimi K2.7 Code
25.8
mirror
Sep 20, 2026
52Qwen3.7 Plus
25.2
mirror
Sep 20, 2026
53Grok 4.3
24.9
mirror
Sep 20, 2026
54Qwen3.5 35B-A3B
24.3
mirror
Aug 7, 2026
55GPT-5.4 mini
24.1
mirror
Sep 20, 2026
56GPT-5.1 Codex
23.7
independent
Sep 9, 2026
57GLM-5V-Turbo
23.5
independent
Sep 9, 2026
58Nemotron 3 Ultra 550B A55B
22.9
mirror
Sep 20, 2026
59MiniMax-M2.5
22.8
independent
Sep 9, 2026
60Gemini 3.5 Flash Lite
22.2
independent
Sep 20, 2026
61Qwen3.6 27B
21.4
mirror
Sep 20, 2026
61Qwen3.5 397B-A17Bnon-reasoning
21.4
independent
Sep 9, 2026
63MiniMax-M2.1
20.9
independent
Sep 9, 2026
64Claude Sonnet 4.5 (latest)
20.7
mirror
Sep 20, 2026
64GPT-5.4 nano
20.7
mirror
Sep 20, 2026
66o3
20.2
independent
Sep 9, 2026
67Qwen3.6 35B-A3B
18.2
mirror
Sep 20, 2026
68Qwen3.5 122B-A10Bnon-reasoning
17.7
independent
Sep 9, 2026
69Kimi K2 Thinkingthinking
17.2
mirror
Aug 7, 2026
70Claude Haiku 4.5 (latest)
16.9
mirror
Sep 20, 2026
70Qwen3-Next 80B-A3B (Thinking)thinking
16.9
mirror
Aug 7, 2026
72GPT-5 Mini
16.8
independent
Sep 20, 2026
73Gemini 2.5 Pro
16.1
mirror
Sep 20, 2026
74Gemini 3.1 Flash Lite Preview
15.6
mirror
Sep 20, 2026
75Gemma 4 31B IT
15.4
mirror
Sep 8, 2026
76GLM-4.7-Flash
14.9
independent
Sep 9, 2026
77GPT-4.1 mini
14.8
mirror
Jul 23, 2026
78Mercury 2
13.8
independent
Sep 20, 2026
79DeepSeek V3.1thinking
13.5
independent
Sep 9, 2026
80Gemini 2.5 Flashthinking
13.1
independent
Sep 9, 2026
81GLM-4.5
12.8
independent
Sep 9, 2026
81Nemotron 3 Super
12.8
mirror
Sep 20, 2026
83GPT-4.1
12.7
independent
Sep 9, 2026
83Qwen3 235B-A22Bthinking
12.7
mirror
Sep 8, 2026
85o3-mini
12.5
independent
Sep 9, 2026
86Qwen3-Coder 480B-A35B Instruct
11.9
independent
Sep 5, 2026
87Sonar Reasoning Pro
11.8
independent
Sep 5, 2026
88DeepSeek R1
11.4
independent
Sep 9, 2026
89GPT-4.1 nano
9.6
mirror
Jul 23, 2026
89Qwen3-Next 80B-A3B Instruct
9.6
independent
Sep 9, 2026
91Llama-3.3-70B-Instruct
9.3
mirror
Aug 7, 2026
92nemotron-3-nano-30b-a3b
8.9
mirror
Sep 8, 2026
93Sonarthinking
8.7
independent
Sep 9, 2026
94Devstral 2
8.6
mirror
Sep 20, 2026
95GLM-4.6V
8.4
independent
Sep 9, 2026
96Magistral Small
8.2
independent
Sep 9, 2026
97GLM-4.5Vthinking
7.6
independent
Sep 9, 2026
97Sonar Pro
7.6
independent
Sep 9, 2026
99Qwen3 32B
7.2
mirror
Sep 8, 2026
100Qwen2.5 32B Instruct
6.9
independent
Sep 9, 2026
101nvidia-nemotron-nano-9b-v2
6.8
independent
Sep 9, 2026
102Gemini 2.5 Flash-Lite
6.7
independent
Sep 9, 2026
103Qwen3 14B
6.4
mirror
Sep 8, 2026
103Qwen Turbo
6.4
independent
Sep 9, 2026
105Qwen3 8B
5.2
mirror
Sep 8, 2026

Bars on a 0–55 index scale. Scores link to model pages; every number keeps its source on the model page.