Skip to content
IFAI

Model rankings

1512 models · 5 boards · 2 sources

‹
Week of September 21, 2026
November 2, 202641 changesOctober 26, 20269 changesOctober 19, 2026247 changesOctober 12, 2026127 changesOctober 5, 202634 changesSeptember 28, 202617 changesSeptember 21, 2026115 changesSeptember 14, 2026174 changesSeptember 7, 2026144 changesAugust 31, 202620 changesAugust 24, 202615 changesAugust 17, 202633 changes
›
Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated September 21, 2026

Smartest
Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic53.4Intelligence
Best open model
GLM-5.3 (max)Z AI44.8Intelligence
Best value
Agnes 3.0 FlashSapiens AI473 pts/$Value
Best from China
Qwen3.8 Max (0902)Alibaba45.4Intelligence
Top three by Intelligence
1Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic53.4
2GPT-6 Astra (max)OpenAI52.7
3Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic50.8
Top three open models
1GLM-5.3 (max)Z AI · #8 overall44.8
2Kimi K3 (max)Kimi · #11 overall43.6
3GLM 5.3 FlashZ AI · #13 overall41.8
This week's changes
claude-opus-5.5-maxCoding · #1New
Gemini 3.8 Flash TTSSpeech · #2New
Grok Imagine Image 2.0Image · #4New
gpt-6-sol-maxCoding · #5New
Gemini 3.8 Flash-Lite TTSSpeech · #6New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
SortIntelligenceValueSpeedLowest price
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1—Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)ClosedAnthropic · 4 more variants53.4$2069242 s
2—GPT-6 Astra (max)ClosedOpenAI · 4 more variants52.7$2069258 s
3—Claude Opus 5 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 4 more variants50.8$106154 s
4—Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)ClosedAnthropic49.6$2065108 s
5—Muse Spark 1.3 (max)ClosedMeta · 1 more variant48.1$2.024124 s
6—GPT-5.6 Sol (max)ClosedOpenAI · 5 more variants47.0$8.069119 s
7▲ 8Qwen3.8 Max (0902)ClosedAlibaba · 1 more variant45.4$3.0373.0 s
8▼ 1GLM-5.3 (max)OpenZ AI44.8$2.2722.9 s
9▼ 1Grok 4.6 (high)ClosedSpaceXAI · 3 more variants44.3$3.06339 s
10NewStep 5 PreviewClosedStepFun43.7$1.41003.0 s
11▼ 2Kimi K3 (max)OpenKimi · 1 more variant43.6$6.0433.7 s
12▼ 2GPT-5.6 Terra (max)ClosedOpenAI · 5 more variants42.1$4.5101200 s
13▼ 1GLM 5.3 FlashOpenZ AI41.8$0.24952.5 s
14▼ 3Claude Opus 4.8 (Adaptive Reasoning, Max Effort)ClosedAnthropic41.8$105828 s
15▼ 2Gemini 3.8 Flash (high)ClosedGoogle · 2 more variants40.9$1.533126 s
16▼ 2Claude Opus 4.7 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 1 more variant40.7$105124 s
17▼ 1Qwen3.8 2.4T A95BOpenAlibaba39.9$3.0403.0 s
18▼ 1Qwen3.8-Flash-NextOpenAlibaba39.8$0.23612.8 s
19—Gemini 3.7 Flash (medium)ClosedGoogle · 2 more variants39.6$1.53034.9 s
20▼ 2Muse Spark 1.2 (xhigh)ClosedMeta39.6$2.023414 s
21▼ 1DeepSeek V4.1 Flash (Reasoning, Max Effort)OpenDeepSeek39.5$0.522290.9 s
22—GPT-5.4 (xhigh)ClosedOpenAI · 2 more variants39.0$5.6140144 s
23▼ 2Grok 4.5 (high)ClosedSpaceXAI38.8$3.0627.1 s
24▼ 1GPT-5.5 (xhigh)ClosedOpenAI · 4 more variants38.4$1111165 s
25▼ 1Claude Sonnet 5 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 5 more variants38.2$4.076132 s
26▼ 1GPT-5.6 Luna (max)ClosedOpenAI · 5 more variants37.3$0.45167127 s
27▼ 1DeepSeek V4 Pro 0813 (Reasoning, Max Effort)OpenDeepSeek · 3 more variants36.0$2.01051.7 s
28▼ 1Agnes 3.0 FlashClosedSapiens AI35.5$0.075——
29▼ 1Agnes 2.5 Pro BetaClosedSapiens AI35.2$0.15——
30▼ 1DeepSeek V4 Flash Vision (Reasoning, Max Effort)ClosedDeepSeek34.8$0.662361.1 s
31▼ 1DeepSeek V4 Flash 0731 (Reasoning, Max Effort)OpenDeepSeek · 3 more variants34.3$0.662441.0 s
32▼ 1Gemini 3.6 Flash (high)ClosedGoogle34.0$1.522333 s
33▼ 1Muse Spark 1.1 (xhigh)ClosedMeta33.7$2.0——
34▼ 1GLM-5.2 (max)OpenZ AI · 1 more variant33.7$2.2765.8 s
35▼ 1Qwen3.8 27B (xhigh)OpenAlibaba · 3 more variants33.7$1.1463.9 s
36▼ 1Gemini 3.5 Flash (medium)ClosedGoogle · 2 more variants33.6$3.424318 s
37▼ 1Motif 3OpenMotif Technologies · 1 more variant33.6———
38▼ 1GPT-5.3 Codex (xhigh)ClosedOpenAI32.5$4.817573 s
39▼ 1Claude Opus 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 1 more variant31.9$104122 s
40—Muse SparkClosedMeta31.3———
41—K2 Horizon 375B A23BOpenInstitute of Foundation Models30.5———
42▲ 1GPT-5.2 (xhigh)ClosedOpenAI · 2 more variants30.4$4.893141 s
43▲ 1Apodex 1.1ClosedApodex30.4$0.98——
44▼ 2Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 2 more variants30.1$6.042152 s
45—Gemini 3.1 Pro PreviewClosedGoogle29.7$4.512464 s
46—Qwen3.7 MaxClosedAlibaba29.5$3.82012.3 s
47—MiniMax-M3OpenMiniMax29.2$0.522270.9 s
48—Claude Opus 4.5 (Reasoning)ClosedAnthropic · 1 more variant29.1$104812 s
49—MiMo-V2-ProClosedXiaomi28.6———
50—GPT-5.2 Codex (xhigh)ClosedOpenAI28.5$4.8——

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 “—”.