Skip to content
IFAI

Model rankings

398 models · 2 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated June 13, 2025

Smartest
o3-proOpenAI71.0Intelligence
Best open model
DeepSeek R1 0528 (May '25)DeepSeek68.3Intelligence
Best value
Gemma 3 4B InstructGoogle979 pts/$Value
Best from China
DeepSeek R1 0528 (May '25)DeepSeek68.3Intelligence
Top three by Intelligence
1o3-proOpenAI71.0
3o3OpenAI70.0
Top three open models
1DeepSeek R1 0528 (May '25)DeepSeek · #5 overall68.3
2Qwen3 235B A22B (Reasoning)Alibaba · #10 overall62.3
This week's changes
o3-proIntelligence · #1New
DeepSeek R1 0528 Qwen3 8BIntelligence · #27New
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1—DeepSeek R1 0528 (May '25)OpenDeepSeek · 1 more variant68.3$0.96——
2—Qwen3 235B A22B (Reasoning)OpenAlibaba · 1 more variant62.3$2.6——
3—Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA60.8$0.90——
4—Qwen3 32B (Reasoning)OpenAlibaba · 1 more variant59.2$2.6——
5—QwQ 32BOpenAlibaba58.1$0.54——
6—Qwen3 14B (Reasoning)OpenAlibaba · 1 more variant55.7$1.3——
7—Qwen3 30B A3B (Reasoning)OpenAlibaba · 1 more variant55.6$0.75——
8—DeepSeek V3 0324 (Mar '25)OpenDeepSeek · 1 more variant53.2$0.48——
9NewDeepSeek R1 0528 Qwen3 8BOpenDeepSeek52.2$0.065——
10▼ 1DeepSeek R1 Distill Qwen 32BOpenDeepSeek51.9$0.30——
11▼ 1Qwen3 8B (Reasoning)OpenAlibaba · 1 more variant51.4$0.66——
12▼ 1Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA · 1 more variant51.2$0.000——
13▼ 1Llama 4 MaverickOpenMeta50.5$0.38——
14▼ 1DeepSeek R1 Distill Qwen 14BOpenDeepSeek49.0$0.20——
15▼ 1DeepSeek R1 Distill Llama 70BOpenDeepSeek48.2$0.72——
16▼ 1Qwen3 4B (Reasoning)OpenAlibaba · 1 more variant47.3$0.40——
17▼ 1Reka Flash 3OpenReka AI47.1$0.35——
18NewLlama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA45.0$0.000——
19▼ 2Llama 4 ScoutOpenMeta43.0$0.24——
20▼ 2QwQ 32B-PreviewOpenAlibaba42.6$0.20——
21▼ 2Llama 3.3 Instruct 70BOpenMeta41.1$0.62——
22▼ 2Llama 3.1 Instruct 405BOpenMeta40.5$3.5——
23▼ 2Qwen2.5 Instruct 72BOpenAlibaba40.4$0.000——
24▼ 2MiniMax-Text-01OpenMiniMax40.2$0.43——
25▼ 2Phi-4OpenMicrosoft Azure40.2$0.22——
26▼ 2Command AOpenCohere40.0$4.4——
27▼ 2Llama 3.1 Tulu3 405BOpenAllen Institute for AI39.9$0.000——
28▼ 2Mistral Large 2 (Nov '24)OpenMistral · 1 more variant38.3$3.0——
29▼ 2Qwen3 1.7B (Reasoning)OpenAlibaba · 1 more variant38.2$0.40——
30▼ 2Gemma 3 27B InstructOpenGoogle37.6$0.000——
31▼ 2Pixtral LargeOpenMistral37.4$3.0——
32▼ 2Qwen2.5 Instruct 32BOpenAlibaba37.4$0.15——
33▼ 2Llama 3.1 Nemotron Instruct 70BOpenNVIDIA37.3$0.16——
34▼ 2Qwen2.5 Coder Instruct 32BOpenAlibaba36.3$0.15——
35▼ 2Llama 3.1 Instruct 70BOpenMeta35.4$0.72——
36▼ 2Mistral Small 3.1OpenMistral35.3$0.15——
37▼ 2Mistral Small 3OpenMistral35.3$0.15——
38▼ 2DeepSeek-V2.5 (Dec '24)OpenDeepSeek · 1 more variant35.3$0.18——
39▼ 2DevstralOpenMistral34.1$0.15——
40▼ 2DeepSeek R1 Distill Llama 8BOpenDeepSeek34.0$0.040——
41▼ 2Gemma 3 12B InstructOpenGoogle33.8$0.063——
42▼ 2R1 1776OpenPerplexity33.7$3.5——
43▼ 2Gemma 3 1B InstructOpenGoogle33.4$0.000——
44▼ 2Llama 3.2 Instruct 90B (Vision)OpenMeta33.4$0.72——
44▼ 2Solar MiniOpenUpstage33.4$0.15——
46▼ 2Qwen2 Instruct 72BOpenAlibaba32.6$0.000——
47▼ 2DeepHermes 3 - Mistral 24B PreviewOpenNous Research30.0$0.000——
48▼ 2Jamba 1.5 LargeOpenAI21 Labs29.3$3.5——
49▼ 2Hermes 3 - Llama-3.1 70BOpenNous Research29.1$0.000——
50▼ 2DeepSeek-Coder-V2OpenDeepSeek29.0$0.18——

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 “—”.