Skip to content
IFAI

Model rankings

526 models · 3 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated September 10, 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
3DeepSeek V3.1 (Reasoning)DeepSeek · #15 overall54.0
This week's changes
Grok Code Fast 1Intelligence · #22New
Qwen3 Max (Preview)Intelligence · #23New
Kimi K2 0905Intelligence · #19Up 12
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1▲ 1gpt-oss-120BOpenOpenAI57.9$0.26——
2▼ 1Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba · 3 more variants57.5$2.6——
3—DeepSeek V3.1 (Reasoning)OpenDeepSeek · 1 more variant54.0$0.96——
4—DeepSeek R1 0528 (May '25)OpenDeepSeek · 1 more variant52.0$0.96——
5▲ 8Kimi K2 0905OpenMoonshot AI · 1 more variant50.4$1.4——
6▼ 1GLM-4.5OpenZ AI49.4$0.99——
7▲ 1GLM-4.5-AirOpenZ AI48.8$0.43——
8▼ 2Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba · 3 more variants46.4$0.75——
9▼ 2MiniMax M1 80kOpenMiniMax46.2$0.83——
10▼ 1Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA · 1 more variant45.2$0.000——
11▲ 1gpt-oss-20BOpenOpenAI44.8$0.088——
12▼ 2Qwen3 4B 2507 (Reasoning)OpenAlibaba · 2 more variants43.4$0.000——
13▼ 2EXAONE 4.0 32B (Reasoning)OpenLG AI Research · 1 more variant42.6$0.70——
14▲ 2Qwen3 Coder 480B A35BOpenAlibaba42.3$3.0——
15—MiniMax M1 40kOpenMiniMax41.6$0.83——
16NewHermes 4 - Llama-3.1 405B (Reasoning)OpenNous Research · 1 more variant41.6$1.5——
17▲ 1DeepSeek V3 0324OpenDeepSeek · 1 more variant41.3$0.48——
18NewHermes 4 - Llama-3.1 70B (Reasoning)OpenNous Research · 1 more variant39.2$0.20——
19▲ 1GLM-4.5V (Reasoning)OpenZ AI · 1 more variant38.8$0.85——
20▼ 3Qwen3 32B (Reasoning)OpenAlibaba · 1 more variant38.7$2.6——
21▼ 7Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA38.5$0.90——
22▼ 3NVIDIA Nemotron Nano 9B V2OpenNVIDIA · 1 more variant38.1$0.000——
23▼ 2QwQ 32BOpenAlibaba37.9$0.63——
24—Qwen3 14B (Reasoning)OpenAlibaba · 1 more variant36.0$1.3——
25▼ 3Llama 4 MaverickOpenMeta35.8$0.39——
26▼ 1Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA · 1 more variant35.5$0.000——
27▼ 1DeepSeek R1 0528 Qwen3 8BOpenDeepSeek34.8$0.065——
28▲ 2Qwen3 Coder 30B A3BOpenAlibaba33.4$0.90——
29▼ 2DeepSeek R1 Distill Qwen 32BOpenDeepSeek32.7$0.20——
30▼ 2Reka Flash 3OpenReka AI32.5$0.35——
31▼ 2Magistral SmallOpenMistral31.9$0.75——
32▼ 1DeepSeek R1 Distill Llama 70BOpenDeepSeek30.8$0.72——
33▼ 1DeepSeek R1 Distill Qwen 14BOpenDeepSeek29.7$0.88——
34—Mistral Small 3.2OpenMistral29.1$0.15——
35▼ 12Qwen3 8B (Reasoning)OpenAlibaba · 1 more variant28.3$0.66——
36▼ 3Llama 4 ScoutOpenMeta28.1$0.26——
37▼ 2Command AOpenCohere28.1$4.4——
38▼ 2QwQ 32B-PreviewOpenAlibaba28.0$0.67——
39—Llama 3.3 Instruct 70BOpenMeta27.9$0.57——
40▼ 3Qwen2.5 Instruct 72BOpenAlibaba27.9$0.000——
41▼ 3Exaone 4.0 1.2B (Reasoning)OpenLG AI Research · 1 more variant26.7$0.000——
42▼ 2Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)OpenNVIDIA26.1$0.000——
43▼ 2MiniMax-Text-01OpenMiniMax25.7$0.43——
44▼ 1Llama 3.1 Instruct 405BOpenMeta25.7$3.8——
45▼ 3Mistral Large 2 (Nov '24)OpenMistral · 1 more variant25.5$3.0——
46▼ 2Llama 3.1 Tulu3 405BOpenAllen Institute for AI25.4$0.000——
47▼ 2Pixtral LargeOpenMistral25.0$3.0——
48▼ 2Grok 2 (Dec '24)OpenxAI24.7$0.000——
49▼ 2Phi-4OpenMicrosoft Azure24.6$0.22——
50▼ 2Mistral Small 3.1OpenMistral23.1$0.15——

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 “—”.