Skip to content
IFAI

Model rankings

761 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

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

Smartest
GPT-5 Codex (high)OpenAI68.5Intelligence
Best open model
Kimi K2 ThinkingMoonshot AI67.0Intelligence
Best value
Gemma 3n E4B InstructGoogle620 pts/$Value
Best from China
Kimi K2 ThinkingMoonshot AI67.0Intelligence
Top three by Intelligence
1GPT-5 Codex (high)OpenAI68.5
2GPT-5 (high)OpenAI68.5
3Kimi K2 ThinkingMoonshot AI67.0
Top three open models
1Kimi K2 ThinkingMoonshot AI · #3 overall67.0
2MiniMax-M2MiniMax · #9 overall61.4
3gpt-oss-120B (high)OpenAI · #10 overall60.5
This week's changes
Kimi K2 ThinkingIntelligence · #3New
GPT-5Image · #5New
Wan 2.5 PreviewImage · #7New
Vivago 2.1Image · #8New
GPT Image 1 (high)Image · #15New
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1NewKimi K2 ThinkingOpenMoonshot AI67.0$1.1——
2▼ 1MiniMax-M2OpenMiniMax61.4$0.52——
3▼ 1DeepSeek V3.1 Terminus (Reasoning)OpenDeepSeek57.7$0.80——
4▼ 1Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba57.5$2.6——
5▼ 1DeepSeek V3.2 Exp (Reasoning)OpenDeepSeek56.9$0.32——
6▼ 1GLM-4.6 (Reasoning)OpenZ AI56.0$1.00——
7▼ 1Qwen3 MaxClosedAlibaba55.2$2.4——
8▼ 1Qwen3 VL 235B A22B (Reasoning)OpenAlibaba54.4$2.6——
9▼ 1Qwen3 Next 80B A3B (Reasoning)OpenAlibaba54.3$1.9——
10▼ 1DeepSeek V3.1 (Reasoning)OpenDeepSeek54.0$0.65——
11▼ 1DeepSeek R1 0528 (May '25)OpenDeepSeek52.0$2.0——
12▼ 1Qwen3 VL 32B (Reasoning)OpenAlibaba51.9$2.6——
13▼ 1Seed-OSS-36B-InstructOpenByteDance Seed51.6$0.30——
14▼ 1GLM-4.5 (Reasoning)OpenZ AI51.3$0.98——
15▼ 1Kimi K2 0905OpenMoonshot AI50.4$1.4——
16▼ 1GLM-4.5-AirOpenZ AI48.8$0.43——
17▼ 1Qwen3 Max (Preview)ClosedAlibaba48.6$2.4——
18▼ 1Kimi K2OpenMoonshot AI48.2$1.1——
19▼ 1Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba46.4$0.75——
20▼ 1DeepSeek V3.2 Exp (Non-reasoning)OpenDeepSeek46.3$0.32——
21▼ 1MiniMax M1 80kOpenMiniMax46.2$0.83——
22▼ 1DeepSeek V3.1 Terminus (Non-reasoning)OpenDeepSeek45.7$0.72——
23▼ 1Qwen3 235B A22B 2507 InstructOpenAlibaba45.3$1.2——
24▼ 1Qwen3 VL 30B A3B (Reasoning)OpenAlibaba45.3$0.75——
25▼ 1Qwen3 Next 80B A3B InstructOpenAlibaba44.8$0.88——
26▼ 1Ling-1TOpenInclusionAI44.8$1.00——
27▼ 1DeepSeek V3.1 (Non-reasoning)OpenDeepSeek44.8$0.84——
28▼ 1GLM-4.6 (Non-reasoning)OpenZ AI44.7$0.99——
29▼ 1Qwen3 VL 235B A22B InstructOpenAlibaba44.2$1.2——
30▼ 1DeepSeek R1 (Jan '25)OpenDeepSeek43.8$2.0——
31▼ 1Qwen3 4B 2507 (Reasoning)OpenAlibaba43.4$0.000——
32▼ 1Qwen3 Coder 480B A35B InstructOpenAlibaba42.3$3.0——
33▼ 1Qwen3 235B A22B (Reasoning)OpenAlibaba41.7$2.6——
34▼ 1DeepSeek V3 0324OpenDeepSeek41.3$1.2——
35▼ 1Qwen3 VL 32B InstructOpenAlibaba41.0$1.2——
36▼ 1MiniMax M1 40kOpenMiniMax40.0$0.83——
37▼ 1Qwen3 Omni 30B A3B (Reasoning)OpenAlibaba40.0$0.43——
38▼ 1Ring-flash-2.0OpenInclusionAI39.5$0.25——
39▼ 1Qwen3 32B (Reasoning)OpenAlibaba38.7$2.6——
40▼ 1Qwen3 VL 30B A3B InstructOpenAlibaba38.5$0.35——
41NewLing-flash-2.0OpenInclusionAI38.3$0.25——
42▼ 2QwQ 32BOpenAlibaba37.9$0.75——
43▼ 2GLM-4.5V (Reasoning)OpenZ AI37.0$0.85——
44▼ 2Qwen3 30B A3B 2507 InstructOpenAlibaba37.0$0.35——
45▼ 2Qwen3 30B A3B (Reasoning)OpenAlibaba36.7$0.75——
46▼ 2Qwen3 14B (Reasoning)OpenAlibaba36.0$1.3——
47▼ 2Qwen3 Coder 30B A3B InstructOpenAlibaba33.4$0.90——
48▼ 2ERNIE 4.5 300B A47BOpenBaidu32.9$0.49——
49▼ 2DeepSeek R1 Distill Qwen 32BOpenDeepSeek32.7$0.29——
50▼ 2DeepSeek V3 (Dec '24)OpenDeepSeek32.5$0.52——

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 “—”.