Skip to main content
Radar

Keep up with the models you depend on. Follow price changes, retirements, and API updates.Follow the models you depend on.

Follow model changes

Artificial Analysis Intelligence Index

A display-only intelligence index published by Artificial Analysis that aggregates provider-reported and benchmark-derived signals into a single model-level score.

Data verified 37 confirmed releases in the last 30 daysSee provider release alerts

The public Artificial Analysis Intelligence Index snapshot ranks GPT-5.6 Sol first at 58.9%, ahead of GPT-5.6 Terra (55.0%) and Claude Fable 5.1 (53.4%) among 204 models. We mirror the table as display-only evidence; it does not affect overall rankings.

Benchmark score on Artificial Analysis Intelligence Index — September 10, 2026

We mirror the published score view for Artificial Analysis Intelligence Index. GPT-5.6 Sol leads the public snapshot at 58.9%, followed by GPT-5.6 Terra (55.0%) and Claude Fable 5.1 (53.4%). We do not use these results to rank models overall.

204 modelsKnowledgeCurrentDisplay onlyUpdated September 10, 2026

Benchmark score table (204 models)

Score
1
GPT-5.6 SolOpenAI · Closed
58.9%
2
GPT-5.6 TerraOpenAI · Closed
55.0%
3
Claude Fable 5.1Anthropic · Closed
53.4%
4
DeepSeek V4 Pro 0813DeepSeek · Closed
53.2%
5
GPT-6 AstraOpenAI · Closed
52.8%
6
GPT-5.6 LunaOpenAI · Closed
51.2%
7
Claude Opus 5Anthropic · Closed
50.7%
8
Gemini 3.5 FlashGoogle · Closed
50.2%
9
Claude Fable 5Anthropic · Closed
49.7%
10
Muse Spark 1.3Meta · Closed
48.2%
11
GLM-5.3Z.AI · Open weight
44.9%
12
Grok 4.6xAI · Closed
44.4%
13
Kimi K3Moonshot AI · Closed
43.8%
14
DeepSeek V4 Pro (High)DeepSeek · Open weight
43.8%
15
Claude Opus 4.8Anthropic · Closed
42.0%
16
GLM-5.3-FlashZ.AI · Open weight
41.9%
17
Hy3 PreviewTencent · Open weight
41.2%
18
Gemini 3.8 FlashGoogle · Closed
41.2%
19
Claude Opus 4.7 (Adaptive)Anthropic · Closed
40.7%
20
Qwen3.8 Max PreviewAlibaba · Closed
40.3%
21
Qwen3.8-Flash-NextAlibaba · Open weight
39.9%
22
Muse Spark 1.2Meta · Closed
39.8%
23
DeepSeek V4.1 FlashDeepSeek · Open weight
39.5%
24
Gemini 3.7 FlashGoogle · Closed
39.4%
25
Grok 4.5xAI · Closed
39.1%
26
GPT-5.4OpenAI · Closed
39.0%
27
GLM-5.2Z.AI · Open weight
38.6%
28
GPT-5.5OpenAI · Closed
38.6%
29
Claude Sonnet 5Anthropic · Closed
38.4%
30
Grok 4.3xAI · Closed
37.6%
31
DeepSeek V4 Flash 0731DeepSeek · Closed
34.5%
32
Gemini 3.6 FlashGoogle · Closed
34.3%
33
Muse Spark 1.1Meta · Closed
34.3%
34
Qwen3.8-27BAlibaba · Open weight
33.9%
35
GPT-5.3 CodexOpenAI · Closed
32.5%
36
GPT-5.3-Codex-SparkOpenAI · Closed
32.5%
37
Claude Opus 4.6 (Adaptive)Anthropic · Closed
31.9%
38
Kimi K2.6Moonshot AI · Open weight
31.3%
39
Muse SparkMeta · Closed
31.3%
40
Claude Opus 4.7Anthropic · Closed
30.9%
41
GPT-5.2OpenAI · Closed
30.4%
42
Apodex 1.1Apodex · Closed
30.4%
43
Apodex 1.1 MiniApodex · Open weight
30.4%
44
Gemini 3.1 ProGoogle · Closed
30.4%
45
Qwen3.7 MaxAlibaba · Closed
29.9%
46
MiniMax M3MiniMax · Open weight
29.6%
47
Claude Opus 4.5 ThinkingAnthropic · Closed
29.1%
48
MiMo-V2-ProXiaomi · Closed
28.6%
49
GPT-5.2-CodexOpenAI · Closed
28.5%
50
Qwen 3.6 Max (preview)Alibaba · Closed
28.4%
51
Solar Pro 4Upstage · Closed
28.1%
52
Gemini 3 ProGoogle · Closed
28.0%
53
GLM-5Z.AI · Open weight
27.9%
54
Quasar 438BMultiverse Computing · Closed
27.1%
55
Qwen3.6 PlusAlibaba · Closed
27.0%
56
GLM-5-TurboZ.AI · Closed
26.6%
57
GLM-5.1Z.AI · Open weight
26.4%
58
MiMo-V2.5-ProXiaomi · Closed
26.4%
59
Claude Opus 4.6Anthropic · Closed
26.4%
60
Kimi K2.7 CodeMoonshot AI · Open weight
26.3%
61
Inkling-SmallThinking Machines Lab · Open weight
26.1%
62
Qwen3.7 PlusAlibaba · Closed
25.8%
63
Hy3Tencent · Open weight
25.8%
64
InklingThinking Machines Lab · Open weight
25.5%
65
Ling 3.0 FlashInclusionAI · Open weight
24.9%
66
Ling 3.0 Flash FP8InclusionAI · Open weight
24.9%
67
Ling 3.0 Flash VLInclusionAI · Open weight
24.8%
68
GPT-5.1OpenAI · Closed
24.7%
69
Claude Sonnet 4.6Anthropic · Closed
24.7%
70
GPT-5.4 miniOpenAI · Closed
24.6%
71
MiMo-V2-OmniXiaomi · Closed
23.9%
72
GPT-5.1-CodexOpenAI · Closed
23.7%
73
GPT-5.1-Codex-MaxOpenAI · Closed
23.7%
74
Claude Opus 4.5Anthropic · Closed
23.7%
75
GLM-5V-TurboZ.AI · Closed
23.5%
76
Kimi K2.5Moonshot AI · Open weight
23.5%
77
Kimi K2.5 (Reasoning)Moonshot AI · Closed
23.5%
78
Nemotron 3 UltraNVIDIA · Open weight
23.4%
79
MiniMax M2.7MiniMax · Open weight
23.2%
80
GPT-5 (high)OpenAI · Closed
23.0%
81
Qwen3.5-27BAlibaba · Open weight
22.9%
82
GPT-5 (medium)OpenAI · Closed
22.9%
83
Claude 4.1 Opus ThinkingAnthropic · Closed
22.9%
84
MiniMax M2.5MiniMax · Closed
22.8%
85
A.X K2SK Telecom · Open weight
22.7%
86
Gemini 3.5 Flash-LiteGoogle · Closed
22.7%
87
Command A+Cohere · Open weight
22.5%
88
Grok 4xAI · Closed
22.5%
89
GLM-4.7Z.AI · Open weight
22.2%
90
Qwen3.6-27BAlibaba · Open weight
21.9%
91
o3-proOpenAI · Closed
21.9%
92
Qwen3.5 397BAlibaba · Open weight
21.4%
93
Qwen3.5 397B (Reasoning)Alibaba · Open weight
21.4%
94
GPT-5.4 nanoOpenAI · Closed
21.2%
95
20.4%
96
o3OpenAI · Closed
20.2%
97
K-EXAONE 2.0LG AI Research · Open weight
19.7%
98
Step 3.7 FlashStepFun · Open weight
19.5%
99
Qwen3.5-35B-A3BAlibaba · Open weight
19.3%
100
Qwen3.6-35B-A3BAlibaba · Open weight
18.8%
101
Claude 4.1 OpusAnthropic · Closed
18.6%
102
Muse Glimmer 30BMeta · Open weight
18.1%
103
17.9%
104
Gemini 3 FlashGoogle · Closed
17.9%
105
GPT-5 miniOpenAI · Closed
17.4%
106
Step 3.5 FlashStepFun · Open weight
17.0%
107
Gemma 4 26B A4BGoogle · Open weight
16.7%
108
Gemini 2.5 ProGoogle · Closed
16.7%
109
Claude 4 SonnetAnthropic · Closed
16.6%
110
Qwen3.5-122B-A10BAlibaba · Open weight
16.2%
111
MiMo-V2-FlashXiaomi · Open weight
16.1%
112
DeepSeek V3.2DeepSeek · Open weight
16.0%
113
Qwen3 MaxAlibaba · Closed
15.6%
114
Gemma 4 31BGoogle · Open weight
15.4%
115
o1OpenAI · Closed
15.2%
116
GLM-4.6Z.AI · Open weight
14.9%
117
Mistral Medium 3.5 128BMistral · Open weight
14.9%
118
GLM-4.7-FlashZ.AI · Open weight
14.9%
119
Granite 4.2 30BIBM · Open weight
14.8%
120
K-ExaoneLG AI Research · Closed
14.4%
121
Gemma 4 12BGoogle · Open weight
14.2%
122
Grok Code Fast 1xAI · Closed
14.1%
123
Ling 2.6 FlashInclusionAI · Open weight
14.1%
124
DeepSeek V3.1DeepSeek · Open weight
13.7%
125
13.6%
126
Nemotron 3 Super 100BNVIDIA · Open weight
13.6%
127
Nemotron 3 Super 120B A12BNVIDIA · Open weight
13.6%
128
DeepSeek V3.1 (Reasoning)DeepSeek · Open weight
13.5%
129
MiniCPM5-2BOpenBMB · Open weight
13.1%
130
DeepSeek-R1DeepSeek · Open weight
13.1%
131
GPT-5 nanoOpenAI · Closed
13.0%
132
North Mini CodeCohere · Open weight
12.8%
133
Kimi K2Moonshot AI · Closed
12.7%
134
GPT-4.1OpenAI · Closed
12.7%
135
o3-miniOpenAI · Closed
12.5%
136
o1-proOpenAI · Closed
12.4%
137
GPT-OSS 120BOpenAI · Open weight
12.3%
138
Ling 3.0 TinyInclusionAI · Open weight
11.9%
139
Granite 4.2 8BIBM · Open weight
11.8%
140
MiniMax M1 80kMiniMax · Closed
11.7%
141
Mercury 2Inception · Closed
11.5%
142
Mistral Small 4Mistral · Open weight
11.4%
143
Mistral Small 4 (Reasoning)Mistral · Open weight
11.4%
144
o1-previewOpenAI · Closed
11.4%
145
Grok 4.1 FastxAI · Closed
11.3%
146
GLM-4.5-AirZ.AI · Closed
11.1%
147
Trinity-Large-ThinkingArcee AI · Open weight
10.9%
148
Trinity-Large-PreviewArcee AI · Open weight
10.9%
149
Nemotron 3 Nano Omni 30B A3BNVIDIA · Open weight
10.3%
150
GPT-4.1 miniOpenAI · Closed
10.2%
151
Gemini 2.5 FlashGoogle · Closed
9.8%
152
Mistral Large 3Mistral · Closed
9.7%
153
Llama 4 MaverickMeta · Open weight
9.3%
154
Granite 4.2 3BIBM · Open weight
9.1%
155
Mistral Medium 3Mistral · Closed
9.1%
156
GPT-OSS 20BOpenAI · Open weight
9.0%
157
Gemma 4 E4BGoogle · Open weight
8.9%
158
Nemotron 3 Nano 30BNVIDIA · Open weight
8.9%
159
Sarvam 105BSarvam · Open weight
8.8%
160
Claude 3 OpusAnthropic · Closed
8.7%
161
DeepSeek V3DeepSeek · Open weight
8.5%
162
GPT-4oOpenAI · Closed
8.4%
163
LFM2.5-2.6BLiquidAI · Open weight
8.4%
164
DeepSeek R1 Distill Qwen 32BDeepSeek · Open weight
8.4%
165
Gemini 1.5 ProGoogle · Closed
7.9%
166
Solar Pro 3Upstage · Closed
7.8%
167
GPT-4.1 nanoOpenAI · Closed
7.8%
168
Gemma 4 E2BGoogle · Open weight
7.8%
169
Qwen3-Omni-30B-A3B-ThinkingAlibaba · Open weight
7.8%
170
Ultravox v0.6 Llama 3.3 70BFixie AI · Open weight
7.7%
171
Mistral Large 2Mistral · Closed
7.6%
172
Nemotron Ultra 253BNVIDIA · Open weight
7.5%
173
Llama 3.1 405BMeta · Open weight
7.3%
174
LFM2.5-8B-A1BLiquidAI · Open weight
7.2%
175
GPT-4 TurboOpenAI · Closed
7.0%
176
Solar Pro 2Upstage · Closed
7.0%
177
Nova ProAmazon · Closed
7.0%
178
Qwen2.5 Coder 32B InstructAlibaba · Open weight
6.7%
179
GPT-4o miniOpenAI · Closed
6.7%
180
Sarvam 30BSarvam · Open weight
6.6%
181
Llama 4 ScoutMeta · Open weight
6.5%
182
Celeris-1Celeris · Closed
6.3%
183
Exaone 4.0 32BLG AI Research · Open weight
6.3%
184
Ministral 3 14B (Reasoning)Mistral · Open weight
6.0%
185
Ministral 3 14BMistral · Open weight
6.0%
186
Qwen3-Omni-30B-A3B-InstructAlibaba · Open weight
6.0%
187
LFM2-24B-A2BLiquidAI · Closed
6.0%
188
Phi-4Microsoft · Open weight
5.9%
189
Phi-4 Multimodal InstructMicrosoft · Open weight
5.8%
190
Claude 3 HaikuAnthropic · Closed
5.6%
191
Ministral 3 8B (Reasoning)Mistral · Open weight
5.5%
192
Ministral 3 8BMistral · Open weight
5.5%
193
Gemini 1.0 ProGoogle · Closed
5.3%
194
Exaone 4.0 1.2BLG AI Research · Open weight
5.2%
195
LFM2.5-1.2B-ThinkingLiquidAI · Closed
5.2%
196
LFM2.5-1.2B-InstructLiquidAI · Closed
5.2%
197
Granite-4.0-H-1BIBM · Open weight
5.2%
198
Granite-4.0-1BIBM · Open weight
5.0%
199
Gemma 3 27BGoogle · Open weight
4.8%
200
Ministral 3 3B (Reasoning)Mistral · Open weight
4.8%
201
Ministral 3 3BMistral · Open weight
4.8%
202
Granite-4.0-350MIBM · Open weight
4.8%
203
Granite-4.0-H-350MIBM · Open weight
4.8%
204
LFM2.5-VL-1.6B-ExtractLiquidAI · Open weight
4.8%

The published Artificial Analysis Intelligence Index snapshot places GPT-5.6 Sol first at 58.9%. The third row is 5.5 points behind. The broader top-10 range is 10.7 points, so the table still separates the published systems.

204 models have been evaluated on Artificial Analysis Intelligence Index. The benchmark falls in the Knowledge category. This category carries a 12% weight in BenchLM.ai's overall scoring system. Artificial Analysis Intelligence Index is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About Artificial Analysis Intelligence Index

Year

2026

Tasks

Cross-benchmark intelligence index

Format

Aggregated model score

Difficulty

Display-only external reference

BenchLM tracks Artificial Analysis as a display-only external reference rather than a weighted benchmark. It is useful as a market snapshot, but it is not a benchmark-native row with a single public task set, scoring harness, or exact-source methodology aligned to BenchLM's core benchmark pages.

BenchLM freshness & provenance

Version

Artificial Analysis Intelligence Index 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does Artificial Analysis Intelligence Index measure?

A display-only intelligence index published by Artificial Analysis that aggregates provider-reported and benchmark-derived signals into a single model-level score.

Which model scores highest on Artificial Analysis Intelligence Index?

GPT-5.6 Sol by OpenAI currently leads with a score of 58.9% on Artificial Analysis Intelligence Index.

How many models are evaluated on Artificial Analysis Intelligence Index?

204 AI models have been evaluated on Artificial Analysis Intelligence Index on BenchLM.

Last updated: September 10, 2026 · BenchLM version Artificial Analysis Intelligence Index 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.