Skip to content
IFAI

Model rankings

526 models · 3 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published September 8, 2025

Most preferred
gemini-2.5-proGoogle1,455Score
Best open model
kimi-k2-0711-previewMoonshot AI1,420Score
Best from China
qwen3-max-previewAlibaba1,428Score
Most voted
claude-3-opus-20240229Anthropic202,641Votes
Top three by Score
1gemini-2.5-proGoogle1,455
3o3-2025-04-16OpenAI1,444
Top three open models
1kimi-k2-0711-previewMoonshot AI · #12 overall1,420
2qwen3-235b-a22b-instruct-2507Alibaba · #13 overall1,420
3deepseek-v3.1-thinkingDeepSeek · #14 overall1,419
This week's changes
Grok Code Fast 1Intelligence · #22New
Qwen3 Max (Preview)Intelligence · #23New
Kimi K2 0905Intelligence · #19Up 12
RankChangeModelScore95% intervalVotesPrice $/M
1Newqwen3-max-previewClosedAlibaba1,428±78,781—
2▼ 1kimi-k2-0711-previewOpenMoonshot AI · 1 more variant1,420±523,527—
3—qwen3-235b-a22b-instruct-2507OpenAlibaba1,420±617,947—
4▼ 2deepseek-v3.1-thinkingOpenDeepSeek · 1 more variant1,419±88,226—
5▼ 1deepseek-r1-0528OpenDeepSeek · 1 more variant1,418±620,863—
6▼ 1glm-4.5OpenZ.ai1,411±614,105—
7—qwen3-235b-a22b-no-thinkingOpenAlibaba1,400±533,139—
8▼ 2qwen3-235b-a22b-thinking-2507OpenAlibaba1,399±78,319—
9▼ 1deepseek-v3-0324OpenDeepSeek · 1 more variant1,391±436,509—
10▼ 1hunyuan-t1-20250711ClosedTencent1,387±85,037—
11—hunyuan-turbos-20250416ClosedTencent · 1 more variant1,380±611,923—
12▼ 2qwen3-coder-480b-a35b-instructOpenAlibaba1,380±615,309—
13▼ 1qwen3-30b-a3b-instruct-2507OpenAlibaba1,380±612,302—
14—qwen3-235b-a22bOpenAlibaba1,373±525,061—
15▼ 2glm-4.5-airOpenZ.ai1,373±613,248—
16▼ 1qwen2.5-maxClosedAlibaba1,372±435,375—
17▼ 1minimax-m1OpenMiniMax1,369±525,260—
18▼ 1step-3OpenStepFun1,350±112,959—
19▼ 1qwen3-32bOpenAlibaba1,346±94,074—
20▼ 1qwen-plus-0125ClosedAlibaba1,344±86,055—
21—glm-4-plus-0111ClosedZhipu · 1 more variant1,342±86,028—
22—hunyuan-turbo-0110ClosedTencent1,340±112,510—
23—qwq-32bOpenAlibaba · 1 more variant1,335±524,230—
24—step-2-16k-exp-202412ClosedStepFun1,332±85,126—
25▲ 1yi-lightningClosed01 AI1,328±528,968—
26▼ 1qwen3-30b-a3bOpenAlibaba1,328±525,172—
27—hunyuan-large-2025-02-10ClosedTencent1,325±103,856—
28—step-1o-turbo-202506ClosedStepFun1,322±610,585—
29—deepseek-v2.5-1210OpenDeepSeek · 1 more variant1,322±87,243—
30—qwen-max-0919OpenAlibaba1,317±517,432—
31—qwen2.5-plus-1127ClosedAlibaba1,315±610,715—
32—hunyuan-standard-2025-02-10ClosedTencent1,309±104,014—
33—qwen2.5-72b-instructOpenAlibaba1,302±441,519—
34—hunyuan-large-visionClosedTencent1,296±96,055—
35—glm-4-0520ClosedZhipu AI1,276±710,221—
36—qwen2.5-coder-32b-instructOpenAlibaba1,272±85,730—
37—deepseek-coder-v2OpenDeepSeek1,265±615,753—
38—qwen2-72b-instructOpenAlibaba1,265±538,872—
39—qwen1.5-110b-chatOpenAlibaba1,237±527,430—
40—hunyuan-standard-256kClosedTencent1,237±112,901—
41—qwen1.5-72b-chatOpenAlibaba1,236±540,658—
42—yi-1.5-34b-chatOpen01 AI1,217±525,135—
43—qwen1.5-32b-chatOpenAlibaba1,209±622,765—
44—qwen1.5-14b-chatOpenAlibaba1,196±718,687—
45—yi-34b-chatOpen01 AI1,188±715,917—
46—deepseek-llm-67b-chatOpenDeepSeek1,187±114,988—
47—qwen1.5-7b-chatOpenAlibaba1,148±104,872—
48—qwen-14b-chatOpenAlibaba1,142±115,065—
49—qwen1.5-4b-chatOpenAlibaba1,095±97,816—

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 “—”.