Skip to content
IFAI

Model rankings

521 models · 3 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated September 1, 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 V3.1 (Reasoning)DeepSeek · #11 overall59.7
This week's changes
Exaone 4.0 1.2B (Reasoning)Intelligence · #81New
Qwen3 4B 2507 (Reasoning)Intelligence · #26Up 59
Qwen3 0.6B (Reasoning)Intelligence · #148Up 10
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1—Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba63.6$2.6——
2—gpt-oss-120BOpenOpenAI61.3$0.26——
3—DeepSeek V3.1 (Reasoning)OpenDeepSeek59.7$0.96——
4—DeepSeek R1 0528 (May '25)OpenDeepSeek58.6$0.96——
5—GLM-4.5OpenZ AI56.1$0.99——
6—Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba53.9$0.75——
7—MiniMax M1 80kOpenMiniMax53.2$0.83——
8—GLM-4.5-AirOpenZ AI52.8$0.43——
9—DeepSeek R1 (Jan '25)OpenDeepSeek52.7$1.8——
10—Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA52.4$0.000——
11NewQwen3 4B 2507 (Reasoning)OpenAlibaba50.8$0.000——
12▼ 1Qwen3 235B A22B 2507 (Non-reasoning)OpenAlibaba50.7$1.2——
13▼ 1EXAONE 4.0 32B (Reasoning)OpenLG AI Research50.7$0.70——
14▼ 1gpt-oss-20BOpenOpenAI49.0$0.088——
15▼ 1DeepSeek V3.1 (Non-reasoning)OpenDeepSeek48.8$0.48——
16▼ 1Kimi K2OpenMoonshot AI48.6$1.1——
17▼ 1Qwen3 235B A22B (Reasoning)OpenAlibaba48.4$2.6——
18▼ 1Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA46.4$0.90——
19▼ 1MiniMax M1 40kOpenMiniMax45.8$0.83——
20▼ 1Qwen3 Coder 480B A35BOpenAlibaba45.2$3.0——
21▼ 1Qwen3 30B A3B 2507 (Non-reasoning)OpenAlibaba44.3$0.35——
22▼ 1Qwen3 32B (Reasoning)OpenAlibaba44.3$2.6——
23▼ 1DeepSeek V3 0324 (Mar '25)OpenDeepSeek44.0$0.48——
24▼ 1NVIDIA Nemotron Nano 9B V2 (Reasoning)OpenNVIDIA43.2$0.000——
25▼ 1GLM-4.5V (Reasoning)OpenZ AI42.9$0.85——
26▼ 1Qwen3 30B A3B (Reasoning)OpenAlibaba42.3$0.75——
27NewNVIDIA Nemotron Nano 9B V2OpenNVIDIA42.1$0.000——
28▼ 2QwQ 32BOpenAlibaba41.9$0.63——
29▼ 2Llama 4 MaverickOpenMeta41.7$0.39——
30▼ 2Qwen3 8B (Reasoning)OpenAlibaba40.8$0.66——
31▼ 2Qwen3 14B (Reasoning)OpenAlibaba40.3$1.3——
32▼ 2Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA39.4$0.000——
33▼ 2DeepSeek R1 0528 Qwen3 8BOpenDeepSeek38.6$0.065——
34▼ 2EXAONE 4.0 32BOpenLG AI Research37.2$0.70——
35▼ 2DeepSeek V3 (Dec '24)OpenDeepSeek36.9$0.48——
36▼ 2DeepSeek R1 Distill Qwen 32BOpenDeepSeek36.4$0.20——
37▼ 2Reka Flash 3OpenReka AI36.3$0.35——
38▼ 2Magistral SmallOpenMistral36.0$0.75——
39▼ 2Qwen3 Coder 30B A3BOpenAlibaba35.6$0.90——
40▼ 2DeepSeek R1 Distill Llama 70BOpenDeepSeek34.5$0.77——
41▼ 1Qwen3 235B A22BOpenAlibaba33.3$1.2——
42▼ 1DeepSeek R1 Distill Qwen 14BOpenDeepSeek33.3$0.88——
43▼ 1Llama 4 ScoutOpenMeta33.1$0.23——
44▼ 1Qwen3 14BOpenAlibaba31.9$0.61——
45▼ 1Mistral Small 3.2OpenMistral31.8$0.15——
46▼ 1Command AOpenCohere31.6$4.4——
47▼ 1QwQ 32B-PreviewOpenAlibaba31.5$0.67——
48▼ 1Qwen2.5 Instruct 72BOpenAlibaba31.4$0.000——
49NewExaone 4.0 1.2B (Reasoning)OpenLG AI Research31.3$0.000——
50▼ 2Llama 3.3 Instruct 70BOpenMeta31.2$0.57——

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 “—”.