Skip to content
IFAI

Model rankings

1396 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated August 3, 2026

Smartest
Best open model
Kimi K3 (max)Kimi57.1Intelligence
Best value
Qwen3.5 4B (Reasoning)Alibaba335 pts/$Value
Best from China
Kimi K3 (max)Kimi57.1Intelligence
Top three open models
1Kimi K3 (max)Kimi · #4 overall57.1
2GLM-5.2 (max)Z AI · #13 overall51.1
This week's changes
muse-spark-1.2 (xHigh)Coding · #13New
gemini-3.5-flash-highCoding · #25New
Inkling SmallIntelligence · #36New
Celeris-1Intelligence · #221New
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1—Claude Opus 5 (Adaptive Reasoning, Max Effort)ClosedAnthropic60.7$105586 s
2—Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)ClosedAnthropic60.1$105541 s
3—Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)ClosedAnthropic59.9$2073131 s
4—GPT-5.6 Sol (max)ClosedOpenAI58.9$1170139 s
5—Claude Opus 5 (Adaptive Reasoning, High Effort)ClosedAnthropic58.9$105319 s
6—GPT-5.6 Sol (xhigh)ClosedOpenAI57.7$116963 s
7▲ 1Claude Opus 5 (Adaptive Reasoning, Medium Effort)ClosedAnthropic56.3$10526.0 s
8▲ 1GPT-5.6 Sol (high)ClosedOpenAI55.9$11609.8 s
9▲ 1Claude Opus 4.8 (Adaptive Reasoning, Max Effort)ClosedAnthropic55.7$105728 s
10▲ 1GPT-5.6 Terra (max)ClosedOpenAI55.0$4.5139164 s
11▲ 1GPT-5.5 (xhigh)ClosedOpenAI54.8$117894 s
12▲ 1Grok 4.5 (high)ClosedSpaceXAI53.8$3.0609.2 s
13▲ 1GPT-5.6 Sol (medium)ClosedOpenAI53.6$11638.9 s
14▲ 1Claude Opus 4.7 (Adaptive Reasoning, Max Effort)ClosedAnthropic53.5$104722 s
15▲ 1Claude Sonnet 5 (Adaptive Reasoning, Max Effort)ClosedAnthropic53.4$4.082202 s
16▲ 1GPT-5.5 (high)ClosedOpenAI53.1$117233 s
17▲ 1GPT-5.6 Terra (xhigh)ClosedOpenAI51.6$4.511320 s
18▲ 1GPT-5.4 (xhigh)ClosedOpenAI51.4$5.6141130 s
19▲ 1GPT-5.6 Luna (max)ClosedOpenAI51.2$0.45181143 s
20▲ 1Muse Spark 1.1 (xhigh)ClosedMeta50.6$2.01802.7 s
21▲ 1Claude Opus 5 (Adaptive Reasoning, Low Effort)ClosedAnthropic50.6$10522.8 s
22▲ 1GPT-5.5 (medium)ClosedOpenAI50.4$11678.2 s
23▲ 1Gemini 3.5 Flash (high)ClosedGoogle50.2$3.418417 s
24▲ 1Gemini 3.6 Flash (high)ClosedGoogle50.1$3.021717 s
25▲ 1GPT-5.6 Sol (low)ClosedOpenAI49.4$11602.9 s
26▲ 1GPT-5.6 Luna (xhigh)ClosedOpenAI49.1$0.4516260 s
27▲ 1GPT-5.6 Terra (high)ClosedOpenAI49.0$4.51122.5 s
28▲ 1Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic47.2$6.055126 s
29▲ 1Gemini 3.1 Pro PreviewClosedGoogle46.5$4.513226 s
30▲ 1GPT-5.6 Luna (high)ClosedOpenAI46.1$0.4515714 s
31▲ 1Qwen3.7 MaxClosedAlibaba46.0$3.82062.4 s
32▲ 1GPT-5.6 Terra (medium)ClosedOpenAI45.6$4.51061.6 s
33▲ 1Gemini 3.5 Flash (medium)ClosedGoogle45.4$3.418018 s
34▲ 1GPT-5.3 Codex (xhigh)ClosedOpenAI44.3$4.813064 s
35▲ 1Motif 3 (Beta)ClosedMotif Technologies44.1———
36▲ 1Claude Opus 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic43.7$104423 s
37▲ 1GPT-5.5 (low)ClosedOpenAI43.5$11691.8 s
38▲ 1Muse SparkClosedMeta43.1———
39▲ 1Claude Opus 4.7 (Non-reasoning, High Effort)ClosedAnthropic42.7$10441.6 s
40▲ 1GPT-5.2 (xhigh)ClosedOpenAI42.2$4.879140 s
41▲ 1Claude Sonnet 5 (Non-reasoning, High Effort)ClosedAnthropic41.7$4.0631.3 s
42▲ 1GPT-5.6 Sol (Non-reasoning)ClosedOpenAI41.2$11641.0 s
43▲ 1Claude Opus 4.5 (Reasoning)ClosedAnthropic40.8$105118 s
44▲ 1GPT-5.6 Terra (low)ClosedOpenAI40.5$4.51031.7 s
45▲ 1MiMo-V2-ProClosedXiaomi40.3———
46▲ 1GPT-5.2 Codex (xhigh)ClosedOpenAI40.1$4.8——
47▲ 1Qwen3.6 Max PreviewClosedAlibaba40.0$2.9543.4 s
48▲ 1GPT-5.4 mini (xhigh)ClosedOpenAI40.0$1.71899.2 s
49▲ 1Grok Build 0.1 0616ClosedSpaceXAI39.8$1.3790.5 s
50▲ 1Qwen3.6 PlusClosedAlibaba39.6$1.1562.1 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 “—”.