Skip to content
IFAI

Model rankings

315 models · 2 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated April 24, 2025

Smartest
o4-mini (high)OpenAI69.8Intelligence
Best open model
Best from China
DeepSeek R1DeepSeek60.2Intelligence
Top three by Intelligence
1o4-mini (high)OpenAI69.8
3o3OpenAI66.8
Top three open models
2DeepSeek R1DeepSeek · #9 overall60.2
3QwQ 32BAlibaba · #10 overall58.1
This week's changes
o4-mini (high)Intelligence · #1New
o3Intelligence · #3New
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1Newo4-mini (high)ClosedOpenAI69.8———
2▼ 1Gemini 2.5 Pro Preview (Mar' 25)ClosedGoogle67.8———
3Newo3ClosedOpenAI66.8———
4▼ 2Grok 3 mini Reasoning (high)ClosedxAI66.7———
5▼ 2o3-mini (high)ClosedOpenAI65.5———
6▼ 2o3-miniClosedOpenAI62.9———
7▼ 2o1ClosedOpenAI61.9———
8NewGemini 2.5 Flash Preview (Reasoning)ClosedGoogle60.5———
9▼ 3Claude 3.7 Sonnet (Extended Thinking)ClosedAnthropic57.4———
10▼ 3o1-miniClosedOpenAI53.8———
11▼ 3GPT-4.1 miniClosedOpenAI52.9———
12▼ 3GPT-4.1ClosedOpenAI52.6———
13▼ 3Gemini 2.0 Flash Thinking Experimental (Jan '25)ClosedGoogle52.3———
14▼ 3Grok 3ClosedxAI50.6———
15▼ 3GPT-4o (March 2025, chatgpt-4o-latest)ClosedOpenAI50.3———
16▼ 3Gemini 2.0 Pro Experimental (Feb '25)ClosedGoogle49.2———
17▼ 3Claude 3.7 Sonnet (Standard)ClosedAnthropic48.2———
18▼ 3Gemini 2.0 Flash (Feb '25)ClosedGoogle48.1———
19▼ 3Gemini 2.0 Flash (experimental)ClosedGoogle46.4———
20▼ 3Qwen2.5 MaxClosedAlibaba45.3———
21▼ 3Gemini 1.5 Pro (Sep '24)ClosedGoogle44.6———
22▼ 3Claude 3.5 Sonnet (Oct '24)ClosedAnthropic44.4———
23▼ 3SonarClosedPerplexity43.4———
24▼ 3Sonar ProClosedPerplexity42.7———
25▼ 3GPT-4o (Nov '24)ClosedOpenAI41.5———
26▼ 3Gemini 2.0 Flash-Lite (Feb '25)ClosedGoogle41.4———
27▼ 3GPT-4.1 nanoClosedOpenAI41.0———
28▼ 3GPT-4o (May '24)ClosedOpenAI40.9———
29▼ 3Grok 2 (Dec '24)ClosedxAI39.2———
30▼ 3Gemini 1.5 Flash (Sep '24)ClosedGoogle39.0———
31▼ 3Grok BetaClosedxAI37.5———
32▼ 3Nova ProClosedAmazon37.1———
33▼ 3GPT-4o miniClosedOpenAI35.7———
34▼ 3Claude 3 OpusClosedAnthropic35.1———
35▼ 3Claude 3.5 HaikuClosedAnthropic34.7———
36▼ 3Gemini 1.5 Pro (May '24)ClosedGoogle33.7———
37▼ 3Qwen TurboClosedAlibaba33.6———
38▼ 3Nova LiteClosedAmazon32.5———
39▼ 3Gemini 1.5 Flash-8BClosedGoogle30.8———
40▼ 3Gemini 1.5 Flash (May '24)ClosedGoogle28.5———
41▼ 3Nova MicroClosedAmazon28.3———
42▼ 3Yi-LargeClosed01.AI27.9———
43▼ 3Claude 3 SonnetClosedAnthropic27.8———
44▼ 3Codestral (Jan '25)ClosedMistral27.6———
45▼ 3Mistral Large (Feb '24)ClosedMistral26.4———
46▼ 3Claude 2.1ClosedAnthropic24.1———
47▼ 3Mistral Small (Feb '24)ClosedMistral22.9———
48▼ 3Mistral MediumClosedMistral22.8———
49▼ 3LFM 40BClosedLiquid AI21.7———
50▼ 3Gemini 1.0 ProClosedGoogle20.6———

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 “—”.