Skip to content
IFAI

Model rankings

532 models · 3 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated September 16, 2025

Smartest
GPT-5 (high)OpenAI66.7Intelligence
Best open model
gpt-oss-120BOpenAI57.9Intelligence
Best value
Gemma 3n E4B InstructGoogle624 pts/$Value
Best from China
Qwen3 235B A22B 2507 (Reasoning)Alibaba57.5Intelligence
Top three by Intelligence
1GPT-5 (high)OpenAI66.7
2Grok 4xAI65.3
3o3-proOpenAI65.3
Top three open models
1gpt-oss-120BOpenAI · #9 overall57.9
2Qwen3 235B A22B 2507 (Reasoning)Alibaba · #10 overall57.5
3Qwen3 Next 80B A3B (Reasoning)Alibaba · #13 overall54.3
This week's changes
Qwen3 Next 80B A3B (Reasoning)Intelligence · #13New
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1—GPT-5 (high)ClosedOpenAI · 3 more variants66.7$3.4——
2—Grok 4ClosedxAI65.3$6.0——
3—o3-proClosedOpenAI65.3$35——
4—o3ClosedOpenAI65.2$3.5——
5—GPT-5 mini (high)ClosedOpenAI · 2 more variants62.3$0.69——
6—Gemini 2.5 ProClosedGoogle59.6$3.4——
7—Claude 4.1 Opus (Extended Thinking)ClosedAnthropic · 1 more variant59.3$30——
8—o4-mini (high)ClosedOpenAI59.0$1.9——
9—gpt-oss-120BOpenOpenAI57.9$0.26——
10—Grok 3 mini Reasoning (high)ClosedxAI57.1$0.35——
11—Claude 4 Sonnet (Extended Thinking)ClosedAnthropic · 1 more variant56.5$6.0——
12—Claude 4 Opus (Extended Thinking)ClosedAnthropic · 1 more variant54.2$30——
13—Gemini 2.5 Pro Preview (Mar' 25)ClosedGoogle · 1 more variant54.1$3.4——
14—Gemini 2.5 Flash (Reasoning)ClosedGoogle · 1 more variant51.2$0.85——
15—o3-mini (high)ClosedOpenAI · 1 more variant50.8$1.9——
16—Grok Code Fast 1ClosedxAI48.7$0.53——
17—GPT-5 nano (high)ClosedOpenAI · 2 more variants48.5$0.14——
18—o1-proClosedOpenAI48.0$263——
19—o1ClosedOpenAI · 1 more variant47.2$26——
20—Sonar Reasoning ProClosedPerplexity46.3$0.000——
21—Gemini 2.5 Flash Preview (Reasoning)ClosedGoogle · 1 more variant45.8$0.000——
22—Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA · 1 more variant45.2$0.000——
23—gpt-oss-20BOpenOpenAI44.8$0.088——
24—GPT-4.1ClosedOpenAI43.4$3.5——
25—Claude 3.7 Sonnet (Extended Thinking)ClosedAnthropic · 1 more variant42.7$6.0——
26—GPT-4.1 miniClosedOpenAI42.5$0.70——
27—Grok 3 Reasoning BetaClosedxAI41.4$0.000——
28—Gemini 2.5 Flash-Lite (Reasoning)ClosedGoogle · 1 more variant40.1$0.18——
29—o1-miniClosedOpenAI39.2$1.9——
30—Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA38.5$0.90——
31—GPT-4.5 (Preview)ClosedOpenAI38.4$0.000——
32—NVIDIA Nemotron Nano 9B V2OpenNVIDIA · 1 more variant38.1$0.000——
33—Gemini 2.0 Flash Thinking Experimental (Jan '25)ClosedGoogle · 1 more variant37.7$0.000——
34—Grok 3ClosedxAI36.0$6.0——
35—Llama 4 MaverickOpenMeta35.8$0.39——
36—GPT-4o (March 2025, chatgpt-4o-latest)ClosedOpenAI · 4 more variants35.6$7.5——
37—Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA · 1 more variant35.5$0.000——
38—Gemini 2.0 Pro Experimental (Feb '25)ClosedGoogle34.6$0.000——
39—Sonar ReasoningClosedPerplexity34.2$2.0——
40—Gemini 2.0 Flash (Feb '25)ClosedGoogle · 1 more variant34.0$0.18——
41—Nova PremierClosedAmazon31.1$5.0——
42—Gemini 1.5 Pro (Sep '24)ClosedGoogle · 1 more variant30.0$0.000——
43—Claude 3.5 Sonnet (Oct '24)ClosedAnthropic · 1 more variant29.9$6.0——
44—SonarClosedPerplexity28.8$1.0——
45—Sonar ProClosedPerplexity28.2$6.0——
46—Llama 4 ScoutOpenMeta28.1$0.26——
47—Llama 3.3 Instruct 70BOpenMeta27.9$0.57——
48—GPT-4.1 nanoClosedOpenAI27.3$0.18——
49—Gemini 2.0 Flash-Lite (Feb '25)ClosedGoogle · 1 more variant26.8$0.13——
50—Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA26.1$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 “—”.