Skip to content
IFAI

Model rankings

340 models · 2 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated May 14, 2025

Smartest
o4-mini (high)OpenAI69.8Intelligence
Best open model
Qwen3 235B A22B (Reasoning)Alibaba62.3Intelligence
Best from China
Qwen3 235B A22B (Reasoning)Alibaba62.3Intelligence
Top three by Intelligence
1o4-mini (high)OpenAI69.8
3o3OpenAI66.8
Top three open models
1Qwen3 235B A22B (Reasoning)Alibaba · #6 overall62.3
3DeepSeek R1DeepSeek · #10 overall60.2
This week's changes
Qwen3 8B (Reasoning)Intelligence · #22New
Mistral Medium 3Intelligence · #29New
Qwen3 4B (Reasoning)Intelligence · #32New
Qwen3 1.7B (Reasoning)Intelligence · #54New
Qwen3 0.6B (Reasoning)Intelligence · #95New
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1—Qwen3 235B A22B (Reasoning)OpenAlibaba62.3———
2—Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA60.8———
3—DeepSeek R1OpenDeepSeek60.2———
4—Qwen3 32B (Reasoning)OpenAlibaba59.2———
5—QwQ 32BOpenAlibaba58.1———
6—Qwen3 14B (Reasoning)OpenAlibaba55.7———
7—Qwen3 30B A3B (Reasoning)OpenAlibaba55.6———
8—DeepSeek V3 0324 (Mar' 25)OpenDeepSeek53.2———
9—DeepSeek R1 Distill Qwen 32BOpenDeepSeek51.9———
10NewQwen3 8B (Reasoning)OpenAlibaba51.4———
11NewLlama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA51.2———
12▼ 2Llama 4 MaverickOpenMeta50.5———
13▼ 2DeepSeek R1 Distill Qwen 14BOpenDeepSeek49.0———
14▼ 2DeepSeek R1 Distill Llama 70BOpenDeepSeek48.2———
15NewQwen3 4B (Reasoning)OpenAlibaba47.3———
16▼ 3Reka Flash 3OpenReka AI47.1———
17NewQwen3 235B A22BOpenAlibaba47.1———
18▼ 4DeepSeek V3 (Dec '24)OpenDeepSeek45.6———
19NewQwen3 32BOpenAlibaba43.5———
20▼ 5Llama 4 ScoutOpenMeta43.0———
21▼ 5QwQ 32B-PreviewOpenAlibaba42.6———
22NewQwen3 30B A3BOpenAlibaba42.6———
23▼ 6Llama 3.3 Instruct 70BOpenMeta41.1———
24NewQwen3 14BOpenAlibaba41.0———
25▼ 7Llama 3.1 Instruct 405BOpenMeta40.5———
26▼ 7Qwen2.5 Instruct 72BOpenAlibaba40.4———
27▼ 7MiniMax-Text-01OpenMiniMax40.2———
28▼ 7Phi-4OpenMicrosoft Azure40.2———
29▼ 7Command AOpenCohere40.0———
30▼ 7Llama 3.1 Tulu3 405BOpenAllen Institute for AI39.9———
31▼ 7Llama 3.3 Nemotron Super 49B v1OpenNVIDIA39.3———
32▼ 7Mistral Large 2 (Nov '24)OpenMistral38.3———
33NewQwen3 1.7B (Reasoning)OpenAlibaba38.2———
34▼ 8Gemma 3 27B InstructOpenGoogle37.6———
35▼ 8Pixtral LargeOpenMistral37.4———
36▼ 8Qwen2.5 Instruct 32BOpenAlibaba37.4———
37▼ 8Llama 3.1 Nemotron Instruct 70BOpenNVIDIA37.3———
38NewQwen3 8BOpenAlibaba36.7———
39▼ 9Mistral Large 2 (Jul '24)OpenMistral36.6———
40▼ 9Qwen2.5 Coder Instruct 32BOpenAlibaba36.3———
41▼ 9Llama 3.1 Instruct 70BOpenMeta35.4———
42▼ 9Mistral Small 3.1OpenMistral35.3———
43▼ 9Mistral Small 3OpenMistral35.3———
44NewQwen3 4BOpenAlibaba35.2———
45▼ 10DeepSeek R1 Distill Llama 8BOpenDeepSeek34.0———
46▼ 10Gemma 3 12B InstructOpenGoogle33.8———
47▼ 10Llama 3.2 Instruct 90B (Vision)OpenMeta33.4———
48▼ 10Qwen2 Instruct 72BOpenAlibaba32.6———
49▼ 10DeepHermes 3 - Mistral 24B PreviewOpenNous Research30.0———
50▼ 10Jamba 1.5 LargeOpenAI21 Labs29.3———

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 “—”.