Skip to content
IFAI

Model rankings

880 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated December 16, 2025

Smartest
Gemini 3 Pro Preview (high)Google72.9Intelligence
Best open model
Kimi K2 ThinkingKimi67.0Intelligence
Best value
Gemma 3n E4B InstructGoogle620 pts/$Value
Best from China
Kimi K2 ThinkingKimi67.0Intelligence
Top three by Intelligence
2GPT-5.2 (xhigh)OpenAI72.6
Top three open models
1Kimi K2 ThinkingKimi · #7 overall67.0
2DeepSeek V3.2 (Reasoning)DeepSeek · #9 overall65.9
3MiniMax-M2MiniMax · #19 overall61.4
This week's changes
GPT-5.2 (xhigh)Intelligence · #2New
gemini-3-flashCoding · #4New
deepseek-v3.2-thinkingCoding · #12New
Apriel-v1.6-15B-ThinkerIntelligence · #31New
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1—Kimi K2 ThinkingOpenKimi67.0$1.1——
2—DeepSeek V3.2 (Reasoning)OpenDeepSeek65.9$0.32——
3—MiniMax-M2OpenMiniMax61.4$0.52——
4—gpt-oss-120B (high)OpenOpenAI60.5$0.26——
5—DeepSeek V3.2 SpecialeOpenDeepSeek58.6$0.32——
6—DeepSeek V3.1 Terminus (Reasoning)OpenDeepSeek57.7$0.80——
7—Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba57.5$2.6——
8NewApriel-v1.6-15B-ThinkerOpenServiceNow57.0$0.000——
9▼ 1DeepSeek V3.2 Exp (Reasoning)OpenDeepSeek56.9$0.32——
10▼ 1GLM-4.6 (Reasoning)OpenZ AI56.0$1.0——
11▼ 1Qwen3 VL 235B A22B (Reasoning)OpenAlibaba54.4$2.6——
12▼ 1Qwen3 Next 80B A3B (Reasoning)OpenAlibaba54.3$1.9——
13▼ 1DeepSeek V3.1 (Reasoning)OpenDeepSeek54.0$0.65——
14▼ 1DeepSeek V3.2 (Non-reasoning)OpenDeepSeek52.4$0.32——
15▼ 1gpt-oss-20B (high)OpenOpenAI52.1$0.10——
16▼ 1DeepSeek R1 0528 (May '25)OpenDeepSeek52.0$1.7——
17▼ 1Qwen3 VL 32B (Reasoning)OpenAlibaba51.9$2.6——
18NewNVIDIA Nemotron 3 Nano 30B A3B (Reasoning)OpenNVIDIA51.8$0.11——
19▼ 2Seed-OSS-36B-InstructOpenByteDance Seed51.6$0.30——
20▼ 2Apriel-v1.5-15B-ThinkerOpenServiceNow51.6$0.000——
21▼ 2GLM-4.5 (Reasoning)OpenZ AI51.3$1.0——
22▼ 2Kimi K2 0905OpenKimi50.4$1.4——
23▼ 2GLM-4.5-AirOpenZ AI48.8$0.43——
24▼ 2Kimi K2OpenKimi48.2$1.1——
25▼ 2gpt-oss-120B (low)OpenOpenAI47.5$0.26——
26▼ 2Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba46.4$0.75——
27▼ 2DeepSeek V3.2 Exp (Non-reasoning)OpenDeepSeek46.3$0.32——
28▼ 2MiniMax M1 80kOpenMiniMax46.2$0.83——
29▼ 2DeepSeek V3.1 Terminus (Non-reasoning)OpenDeepSeek45.7$0.72——
30▼ 2Qwen3 235B A22B 2507 InstructOpenAlibaba45.3$1.2——
31▼ 2Qwen3 VL 30B A3B (Reasoning)OpenAlibaba45.3$0.75——
32▼ 2Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA45.2$0.18——
33▼ 2Qwen3 Next 80B A3B InstructOpenAlibaba44.8$0.88——
34▼ 2Ling-1TOpenInclusionAI44.8$1.00——
35▼ 2DeepSeek V3.1 (Non-reasoning)OpenDeepSeek44.8$0.84——
36▼ 2GLM-4.6 (Non-reasoning)OpenZ AI44.7$1.0——
37▼ 2gpt-oss-20B (low)OpenOpenAI44.3$0.10——
38▼ 2Qwen3 VL 235B A22B InstructOpenAlibaba44.2$1.2——
39▼ 2DeepSeek R1 (Jan '25)OpenDeepSeek43.8$2.0——
40▼ 2Qwen3 4B 2507 (Reasoning)OpenAlibaba43.4$0.000——
41▼ 2Magistral Small 1.2OpenMistral43.0$0.75——
42▼ 2EXAONE 4.0 32B (Reasoning)OpenLG AI Research42.6$0.70——
43▼ 2Qwen3 Coder 480B A35B InstructOpenAlibaba42.3$3.0——
44▼ 2Ring-1TOpenInclusionAI41.8$0.99——
45▼ 2Qwen3 235B A22B (Reasoning)OpenAlibaba41.7$2.6——
46▼ 2Hermes 4 - Llama-3.1 405B (Reasoning)OpenNous Research41.6$1.5——
47▼ 2DeepSeek V3 0324OpenDeepSeek41.3$1.2——
48▼ 2Qwen3 VL 32B InstructOpenAlibaba41.0$1.2——
49▼ 2NVIDIA Nemotron Nano 12B v2 VL (Reasoning)OpenNVIDIA40.7$0.30——
50▼ 2MiniMax M1 40kOpenMiniMax40.0$0.83——

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 “—”.