Skip to content
IFAI

Model rankings

1388 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated July 27, 2026

Smartest
Best open model
GLM-5.2 (max)Z AI51.1Intelligence
Best value
Qwen3.5 4B (Reasoning)Alibaba335 pts/$Value
Best from China
Kimi K3Kimi57.1Intelligence
Top three open models
1GLM-5.2 (max)Z AI · #13 overall51.1
2MiniMax-M3MiniMax · #20 overall44.4
3DeepSeek V4 Pro (Reasoning, Max Effort)DeepSeek · #21 overall44.3
This week's changes
claude-opus-5-maxCoding · #1New
Reve 2.1Image · #2New
MiniMax H3Video · #2New
qwen3.8-maxCoding · #4New
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1NewClaude Opus 5 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 4 more variants60.7$105370 s
2▼ 1Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)ClosedAnthropic59.9$207099 s
3▼ 1GPT-5.6 Sol (max)ClosedOpenAI · 5 more variants58.9$1177124 s
4▼ 1Kimi K3ClosedKimi57.1$6.032165 s
5▼ 1Claude Opus 4.8 (Adaptive Reasoning, Max Effort)ClosedAnthropic55.7$105529 s
6▼ 1GPT-5.6 Terra (max)ClosedOpenAI · 5 more variants55.0$5.6134164 s
7▼ 1GPT-5.5 (xhigh)ClosedOpenAI · 4 more variants54.8$117264 s
8▼ 1Grok 4.5 (high)ClosedSpaceXAI53.8$3.0588.3 s
9▼ 1Claude Opus 4.7 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 1 more variant53.5$104523 s
10▼ 1Claude Sonnet 5 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 1 more variant53.4$4.075193 s
11▼ 1GPT-5.4 (xhigh)ClosedOpenAI · 2 more variants51.4$5.6144127 s
12▼ 1GPT-5.6 Luna (max)ClosedOpenAI · 5 more variants51.2$2.3188117 s
13▼ 1GLM-5.2 (max)OpenZ AI · 1 more variant51.1$2.22171.3 s
14▼ 1Muse Spark 1.1 (xhigh)ClosedMeta50.6$2.01261.4 s
15▼ 1Gemini 3.5 Flash (high)ClosedGoogle · 2 more variants50.2$3.418621 s
16NewGemini 3.6 Flash (high)ClosedGoogle50.1$3.023517 s
17▼ 2Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 2 more variants47.2$6.046116 s
18▼ 2Gemini 3.1 Pro PreviewClosedGoogle46.5$4.512542 s
19▼ 2Qwen3.7 MaxClosedAlibaba46.0$3.82022.6 s
20▼ 2MiniMax-M3OpenMiniMax44.4$0.52731.5 s
21▼ 2DeepSeek V4 Pro (Reasoning, Max Effort)OpenDeepSeek · 2 more variants44.3$0.54711.5 s
22▼ 2GPT-5.3 Codex (xhigh)ClosedOpenAI44.3$4.814363 s
23▼ 2Kimi K2.6OpenKimi · 1 more variant44.2$1.7433.0 s
24NewMotif 3 (Beta)ClosedMotif Technologies44.1———
25▼ 3Claude Opus 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 1 more variant43.7$103814 s
26▼ 3Muse SparkClosedMeta43.1———
27▼ 3MiMo-V2.5-ProOpenXiaomi · 1 more variant42.2$0.54713.2 s
28▼ 3GPT-5.2 (xhigh)ClosedOpenAI · 2 more variants42.2$4.879112 s
29▼ 3Kimi K2.7 CodeOpenKimi41.9$1.7482.8 s
30▼ 3Hy3OpenTencent41.2$0.25562.6 s
31▼ 3Nex-N2-ProOpenNex AGI41.0$1.01411.8 s
32▼ 3Claude Opus 4.5 (Reasoning)ClosedAnthropic · 1 more variant40.8$104713 s
33▼ 3Inkling (xhigh)OpenThinking Machines40.7$2.6881.8 s
34▼ 3MiMo-V2-ProClosedXiaomi40.3———
35▼ 3DeepSeek V4 Flash (Reasoning, Max Effort)OpenDeepSeek · 2 more variants40.3$0.181151.2 s
36▼ 3GLM-5.1 (Reasoning)OpenZ AI · 1 more variant40.2$2.1721.8 s
37▼ 3GPT-5.2 Codex (xhigh)ClosedOpenAI40.1$4.8——
38▼ 3Qwen3.6 Max PreviewClosedAlibaba40.0$2.9533.2 s
39▼ 3GPT-5.4 mini (xhigh)ClosedOpenAI · 2 more variants40.0$1.71658.0 s
40▼ 3Grok Build 0.1 0616ClosedSpaceXAI39.8$1.3840.5 s
41▼ 3Qwen3.6 PlusClosedAlibaba39.6$1.1532.5 s
42▼ 3Gemini 3 Pro Preview (high)ClosedGoogle · 1 more variant39.6$4.5——
43▼ 3GLM-5 (Reasoning)OpenZ AI · 1 more variant39.5$1.6431.6 s
44▼ 3Qwen3.7 PlusClosedAlibaba39.0$0.70542.8 s
45▼ 3JT-4.1 Flash 236B A21BClosedChina Mobile38.8———
46NewAgnes 2.5 Pro AlphaClosedSapiens AI38.8$0.561282.0 s
47▼ 4GPT-5.4 nano (xhigh)ClosedOpenAI · 2 more variants38.2$0.461564.5 s
48▼ 4MiniMax-M2.7OpenMiniMax38.1$0.52561.8 s
49▼ 4GLM-5-TurboClosedZ AI38.1———
50▼ 4Nemotron 3 Ultra 550B A55B (Reasoning)OpenNVIDIA37.8$1.21961.2 s

Intelligence over time

Each maker's best model at the time on the Artificial Analysis Intelligence Index, by the month it was released.

Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi

Arena data comes from the leaderboard dataset it publishes on Hugging Face (CC BY 4.0). Only each board's overall leaderboard is used; scores and ranks are unchanged. dataset · CC BY 4.0

Artificial Analysis data comes from its API (artificialanalysis.ai). Scores are unchanged; ranks follow the scores. artificialanalysis.ai

Open or closed follows the license Arena gives; prices and Artificial Analysis intelligence show only where model names match, otherwise “—”.