Skip to content
IFAI

Model rankings

779 models · 5 boards · 2 sources

‹
Week of November 17, 2025
December 29, 202513 changesDecember 22, 20256 changesDecember 15, 20258 changesDecember 8, 202520 changesDecember 1, 202522 changesNovember 24, 202546 changesNovember 17, 202514 changesNovember 10, 202510 changesNovember 3, 202523 changesOctober 27, 20253 changesOctober 20, 20252 changesOctober 13, 202518 changes
›
Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published November 20, 2025

Most preferred
gemini-3-proGoogle1,495Score
Best open model
kimi-k2-thinking-turboMoonshot AI1,429Score
Best from China
qwen3-max-previewAlibaba1,433Score
Most voted
claude-3-opus-20240229Anthropic196,368Votes
Top three by Score
1gemini-3-proGoogle1,495
2grok-4.1-thinkingxAI1,481
3gpt-5.1-highOpenAI1,454
Top three open models
1kimi-k2-thinking-turboMoonshot AI · #15 overall1,429
2glm-4.6Z.ai · #16 overall1,426
3qwen3-235b-a22b-instruct-2507Alibaba · #20 overall1,421
This week's changes
GPT-5.1 (high)Intelligence · #1New
Inworld TTS 1 MaxSpeech · #1New
gpt-5.1-mediumCoding · #2New
gpt-5.1-codexCoding · #10New
gpt-5.1-codex-miniCoding · #14New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
SortScoreLowest price
RankChangeModelScore95% intervalVotesPrice $/M
1—kimi-k2-thinking-turboOpenMoonshot AI1,429±78,126—
2▲ 1glm-4.6OpenZ.ai1,426±516,268—
3▲ 1qwen3-235b-a22b-instruct-2507OpenAlibaba1,421±442,468—
4▼ 2deepseek-v3.2-exp-thinkingOpenDeepSeek · 1 more variant1,421±79,228—
5—deepseek-r1-0528OpenDeepSeek · 1 more variant1,418±619,246—
6—kimi-k2-0905-previewOpenMoonshot AI · 1 more variant1,416±710,652—
7—deepseek-v3.1OpenDeepSeek · 1 more variant1,416±615,261—
8—deepseek-v3.1-terminusOpenDeepSeek · 1 more variant1,415±103,751—
9—qwen3-vl-235b-a22b-instructOpenAlibaba1,415±78,539—
10—glm-4.5OpenZ.ai1,408±524,834—
11—qwen3-next-80b-a3b-instructOpenAlibaba1,400±523,138—
12—longcat-flash-chatOpenMeituan1,399±611,509—
13—qwen3-235b-a22b-no-thinkingOpenAlibaba1,399±539,395—
14—qwen3-235b-a22b-thinking-2507OpenAlibaba1,397±69,349—
15—qwen3-vl-235b-a22b-thinkingOpenAlibaba1,393±77,991—
16—deepseek-v3-0324OpenDeepSeek · 1 more variant1,391±446,819—
17—qwen3-coder-480b-a35b-instructOpenAlibaba1,385±523,162—
18—qwen3-30b-a3b-instruct-2507OpenAlibaba1,382±524,219—
19—qwen3-235b-a22bOpenAlibaba1,374±527,179—
20—glm-4.5-airOpenZ.ai1,370±431,688—
21—qwen3-next-80b-a3b-thinkingOpenAlibaba1,367±613,838—
22—minimax-m1OpenMiniMax1,365±436,890—
23—gemma-3-27b-itOpenGoogle1,364±449,341—
24—mistral-small-2506OpenMistral1,354±518,337—
25—gpt-oss-120bOpenOpenAI1,352±431,298$0.26
26—command-a-03-2025OpenCohere1,352±357,871—
27—glm-4.5vOpenZ.ai1,352±84,984—
28—minimax-m2OpenMiniMax1,346±87,132—
29—ling-flash-2.0OpenAnt Group1,346±77,160—
30—llama-3.1-nemotron-ultra-253b-v1OpenNVIDIA1,345±122,573—
31—step-3OpenStepFun1,345±76,652—
32—qwen3-32bOpenAlibaba1,344±93,943—
33—gemma-3-12b-itOpenGoogle1,340±93,866—
34—nvidia-llama-3.3-nemotron-super-49b-v1.5OpenNVIDIA1,340±103,496—
35—llama-3.1-405b-instruct-bf16OpenMeta1,335±441,932—
36▲ 1llama-3.1-405b-instruct-fp8OpenMeta1,334±360,272—
37▼ 1qwq-32bOpenAlibaba · 1 more variant1,333±426,277—
38—llama-4-maverick-17b-128e-instructOpenMeta1,327±441,227—
39—qwen3-30b-a3bOpenAlibaba1,326±527,509—
40—llama-3.3-nemotron-49b-super-v1OpenNVIDIA1,325±122,243—
41—llama-4-scout-17b-16e-instructOpenMeta1,322±531,219—
42—deepseek-v2.5-1210OpenDeepSeek · 1 more variant1,322±86,877—
43—ring-flash-2.0OpenAnt Group1,319±77,286—
44—llama-3.3-70b-instructOpenMeta1,319±356,021—
45—gemma-3n-e4b-itOpenGoogle1,318±523,483—
46—gpt-oss-20bOpenOpenAI1,318±610,858—
47—qwen-max-0919OpenAlibaba1,317±616,598—
48▲ 1mistral-large-2407OpenMistral · 1 more variant1,313±445,968—
49▼ 1athene-v2-chatOpenNexusFlow1,313±424,880—
50—athene-70b-0725OpenNexusFlow1,305±619,796—

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 “—”.