/rankings
Rankings
97 models · ranked by Artificial Analysis Intelligence Index
#
Model
Score
Context
1
GLM 5.1
z-ai
51.4
202K
→
2
GLM 5
z-ai
49.8
202K
→
3
MiniMax M2.7
minimax
49.6
196K
→
4
Kimi K2.5
moonshotai
46.8
262K
→
5
Gemini 3 Flash
google
46
1000M
→
6
Qwen3.5 397B VLM (a17b)
qwen
45
131.1M
→
7
GPT-5
openai
44.6
400K
→
8
GLM 4.7
z-ai
42.1
202K
→
9
Qwen3.5 122B (a10b)
qwen
42
131.1M
→
10
MiniMax M2.5
minimax
41.9
196K
→
11
DeepSeek V3.2
deepseek
41.7
131K
→
12
GPT-5-mini
openai
41.2
400K
→
13
Gemma 4 31B IT
google
39.2
262K
→
14
Qwen3.5 4B
qwen
39
32.8M
→
15
o3
openai
38.4
200K
→
16
GPT-4o mini
openai
38
128K
→
17
Step 3.5 Flash 196B
stepfun
38
128M
→
18
Kimi K2 Instruct
moonshotai
37
128M
→
19
Nemotron 3 Super 120B
nvidia
36
262K
→
20
GPT-5-chat (preview)
openai
35
128K
→
21
Gemini 3.1 Flash-Lite
google
33.5
128M
→
22
gpt-oss-120b
openai
33.3
131K
→
23
o4-mini
openai
33.1
200K
→
24
Qwen3.5 9B
qwen
32.4
262K
→
25
Grok 3
x-ai
32
131K
→
26
DeepSeek-V3-0324
deepseek
32
128M
→
27
Grok 3 Mini
x-ai
32
131K
→
28
Gemma 4 26B A4B IT
google
31.2
262K
→
29
o1
openai
30.8
200K
→
30
MiMo V2 Flash 309B
xiaomi
30
131.1M
→
31
DeepSeek V3.1
deepseek
28.1
32K
→
32
Mistral Small 4 119B
mistralai
28
128M
→
33
DeepSeek-R1-0528
deepseek
27.1
163K
→
34
DeepSeek-R1-0528-Qwen3-8B
deepseek
27
128K
→
35
GPT-5-nano
openai
26.8
400K
→
36
Qwen3-Next 80B A3B Thinking
qwen
26.7
131K
→
37
Kimi K2
moonshotai
26.3
131K
→
38
GPT-4.1
openai
26.3
1.0M
→
39
Upstage Solar Pro 3
upstage
25.9
128K
→
40
o3-mini
openai
25.2
200K
→
41
GPT-4o
openai
25
128K
→
42
gpt-oss-20b
openai
24.5
131K
→
43
Nemotron 3 Nano 30B A3B
nvidia
24.3
262K
→
44
gpt-oss-safeguard-20b
openai
24
131K
→
45
Mistral Large 3
mistralai
23
128M
→
46
GPT-4.1-mini
openai
22.9
1.0M
→
47
Devstral 2 123B Instruct
mistralai
22
128M
→
48
Mistral Nemo
mistralai
21
131K
→
49
Mistral Medium 3
mistralai
21
128M
→
50
Gemini 2.5 Flash
google
20.6
1.0M
→
51
Qwen3-Next 80B A3B Instruct
qwen
20.1
262K
→
52
Qwen3 235B A22B Instruct
qwen
20
131.1M
→
53
Devstral Small / Medium
mistralai
19.5
131K
→
54
DeepSeek-R1
deepseek
18.8
64K
→
55
Llama 4 Maverick 17B 128E
meta-llama
18
1000M
→
56
Magistral Small / Medium
mistralai
18
128M
→
57
Qwen 2.5 72B Instruct
qwen
18
32K
→
58
Gemini 2.5 Flash-Lite
google
17.6
1.0M
→
59
Llama 3.1 405B Instruct
meta-llama
17.4
128K
→
60
Qwen3 32B
qwen
16.5
40K
→
61
DeepSeek-R1-Distill-Llama-70B
deepseek
16
131K
→
62
Qwen3 4B
qwen
16
32K
→
63
Ministral 3B
mistralai
16
128K
→
64
Falcon 7B Instruct
tii
16
2.0M
→
65
Ministral 8B
mistralai
15
128K
→
66
Arcee Trinity Mini
arcee-ai
15
131K
→
67
Nemotron Nano 12B V2 VL
nvidia
14.9
131K
→
68
Nemotron Nano 9B V2
nvidia
14.8
131K
→
69
Llama 3.3 70B Instruct
meta-llama
14.5
131K
→
70
Mistral Small 3.1 24B Instruct
mistralai
14.5
128K
→
71
Llama 4 Scout 17B 16E
meta-llama
14
1000M
→
72
Command A
cohere
13.5
256K
→
73
Qwen3 8B
qwen
13.2
40K
→
74
GPT-4.1-nano
openai
13
1.0M
→
75
Llama 3.1 70B Instruct
meta-llama
12.5
131K
→
76
Llama 3.2 90B Vision
meta-llama
12
128K
→
77
Qwen 2.5 7B Instruct
qwen
12
32K
→
78
Llama 3.1 8B Instruct
meta-llama
11.8
16K
→
79
IBM Granite 4.0 H Micro
ibm
11
8.2M
→
80
Qwen 2.5 Coder 7B Instruct
qwen
11
131.1M
→
81
Phi-4 Reasoning
microsoft
10.4
32K
→
82
Gemma 3 27B IT
google
10.3
131K
→
83
Phi-4 Multimodal Instruct
microsoft
10
128K
→
84
Phi-4
microsoft
10
16K
→
85
AI21 Jamba 1.5 Large
ai21
9
256M
→
86
Gemma 3 12B IT
google
8.8
131K
→
87
Llama 3.2 11B Vision
meta-llama
8.7
131K
→
88
Phi-4 Mini Instruct
microsoft
8.4
128M
→
89
Gemma 7B IT
google
8
8K
→
90
Gemma 3n e4b IT
google
6.4
32K
→
91
Llama 3 8B Instruct
meta-llama
6.4
8K
→
92
Gemma 3 4B IT
google
6.3
131K
→
93
Llama 3.2 1B Instruct
meta-llama
6.3
60K
→
94
Llama 3.2 3B Instruct
meta-llama
6
80K
→
95
Gemma 3 1B IT
google
5.5
32K
→
96
Tiny Aya Global
cohere
5
8.2M
→
97
Gemma 3n e2b IT
google
4.8
32K
→
Scores from Artificial Analysis. Higher is better.