Skip to content
IFAI

Model rankings

505 models · 3 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated August 18, 2025

Smartest
GPT-5 (high)OpenAI69.0Intelligence
Best open model
Qwen3 235B A22B 2507 (Reasoning)Alibaba63.6Intelligence
Best value
Gemma 3n E4B InstructGoogle739 pts/$Value
Best from China
Qwen3 235B A22B 2507 (Reasoning)Alibaba63.6Intelligence
Top three by Intelligence
1GPT-5 (high)OpenAI69.0
2Grok 4xAI67.5
3o3-proOpenAI67.5
Top three open models
1Qwen3 235B A22B 2507 (Reasoning)Alibaba · #8 overall63.6
2gpt-oss-120BOpenAI · #9 overall61.3
3DeepSeek R1 0528 (May '25)DeepSeek · #13 overall58.7
This week's changes
Mistral Medium 3.1Intelligence · #58New
Gemma 3 4B InstructIntelligence · #132Up 13
Gemini 2.5 Flash-Lite (Reasoning)Intelligence · #32Up 7
GPT-4.1 miniIntelligence · #38Up 7
GPT-4.1 nanoIntelligence · #76Up 3
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1—GPT-5 (high)ClosedOpenAI · 3 more variants69.0$3.4——
2—Grok 4ClosedxAI67.5$6.0——
3—o3-proClosedOpenAI67.5$35——
4—o3ClosedOpenAI67.1$3.5——
5—o4-mini (high)ClosedOpenAI65.1$1.9——
6—Gemini 2.5 ProClosedGoogle64.6$3.4——
7—GPT-5 miniClosedOpenAI63.7$0.69——
8—Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba · 3 more variants63.6$2.6——
9▲ 1gpt-oss-120BOpenOpenAI61.3$0.26——
10▼ 1Claude 4.1 Opus (Extended Thinking)ClosedAnthropic · 1 more variant61.0$30——
11—Gemini 2.5 Pro Preview (Mar' 25)ClosedGoogle · 1 more variant59.0$3.4——
12—Claude 4 Sonnet (Extended Thinking)ClosedAnthropic · 1 more variant58.8$6.0——
13—DeepSeek R1 0528 (May '25)OpenDeepSeek · 1 more variant58.7$0.96——
14—Gemini 2.5 Flash (Reasoning)ClosedGoogle · 1 more variant58.4$0.85——
15—Grok 3 mini Reasoning (high)ClosedxAI58.0$0.35——
16—GLM-4.5OpenZ AI56.1$0.96——
17—o3-mini (high)ClosedOpenAI · 1 more variant55.5$1.9——
18—Claude 4 Opus (Extended Thinking)ClosedAnthropic · 1 more variant55.3$30——
19—GPT-5 nanoClosedOpenAI53.8$0.14——
20—Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba · 3 more variants53.2$0.75——
21—MiniMax M1 80kOpenMiniMax53.2$0.83——
22—Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA · 1 more variant52.4$0.000——
23—o1ClosedOpenAI · 1 more variant51.7$26——
24—MiniMax M1 40kOpenMiniMax51.1$0.83——
25—Sonar Reasoning ProClosedPerplexity50.7$0.000——
26—EXAONE 4.0 32B (Reasoning)OpenLG AI Research · 1 more variant50.7$0.70——
27—Gemini 2.5 Flash Preview (Reasoning)ClosedGoogle · 1 more variant50.2$0.99——
28—GLM-4.5-AirOpenZ AI49.5$0.43——
29—gpt-oss-20BOpenOpenAI49.0$0.088——
30—Kimi K2OpenMoonshot AI48.6$1.1——
31—QwQ 32BOpenAlibaba47.7$0.47——
32▲ 7Gemini 2.5 Flash-Lite (Reasoning)ClosedGoogle · 1 more variant47.3$0.18——
33▼ 1Claude 3.7 Sonnet (Extended Thinking)ClosedAnthropic · 1 more variant47.0$6.0——
34▼ 1GPT-4.1ClosedOpenAI46.8$3.5——
35▼ 1Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA46.4$0.90——
36▼ 1Grok 3 Reasoning BetaClosedxAI45.6$0.000——
36▼ 1o1-proClosedOpenAI45.6$263——
38▲ 7GPT-4.1 miniClosedOpenAI45.6$0.70——
39▼ 1Qwen3 Coder 480B A35BOpenAlibaba45.2$3.0——
40—Qwen3 32B (Reasoning)OpenAlibaba · 1 more variant44.3$2.6——
41—DeepSeek V3 0324 (Mar '25)OpenDeepSeek · 1 more variant44.0$0.48——
42—Solar Pro 2 (Reasoning)ClosedUpstage · 3 more variants43.3$0.50——
43—o1-miniClosedOpenAI43.3$1.9——
44—GPT-4.5 (Preview)ClosedOpenAI42.4$0.000——
45▲ 1Llama 4 MaverickOpenMeta41.7$0.39——
46▲ 1Gemini 2.0 Flash Thinking Experimental (Jan '25)ClosedGoogle · 1 more variant41.7$0.000——
47▲ 1DeepSeek R1 0528 Qwen3 8BOpenDeepSeek41.5$0.065——
48▲ 1DeepSeek R1 Distill Qwen 32BOpenDeepSeek41.2$0.20——
49▲ 1Qwen3 8B (Reasoning)OpenAlibaba · 1 more variant40.8$0.66——
50▲ 1Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA · 1 more variant40.5$0.000——

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 “—”.