Skip to content
IFAI

Model rankings

740 models · 4 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published November 9, 2025

Most preferred
gemini-2.5-proGoogle1,452Score
Best open model
glm-4.6Z.ai1,429Score
Best from China
qwen3-max-previewAlibaba1,432Score
Most voted
claude-3-opus-20240229Anthropic196,368Votes
Top three by Score
1gemini-2.5-proGoogle1,452
Top three open models
1glm-4.6Z.ai · #11 overall1,429
2deepseek-v3.2-expDeepSeek · #15 overall1,422
3qwen3-235b-a22b-instruct-2507Alibaba · #17 overall1,420
This week's changes
MiniMax-M2Intelligence · #8New
Chirp 3: HDSpeech · #20New
Maya1Speech · #21New
Qwen3 VL 235B A22B (Reasoning)Intelligence · #23New
Sonic 3Speech · #23New
RankChangeModelScore95% intervalVotesPrice $/M
1—qwen3-max-previewClosedAlibaba1,432±524,709—
2▲ 1glm-4.6OpenZ.ai1,429±610,168—
3▼ 1qwen3-max-2025-09-23ClosedAlibaba1,423±69,298—
4—deepseek-v3.2-expOpenDeepSeek · 1 more variant1,422±27434—
5▲ 1qwen3-235b-a22b-instruct-2507OpenAlibaba1,420±436,122—
6▲ 1deepseek-r1-0528OpenDeepSeek · 1 more variant1,417±619,266—
7▲ 1kimi-k2-0905-previewOpenMoonshot AI · 1 more variant1,417±710,705—
8▲ 1deepseek-v3.1OpenDeepSeek · 1 more variant1,416±615,303—
9▼ 4qwen3-vl-235b-a22b-instructOpenAlibaba1,415±78,589—
10—deepseek-v3.1-terminusOpenDeepSeek · 1 more variant1,415±103,763—
11—glm-4.5OpenZ.ai1,408±524,926—
12—qwen3-next-80b-a3b-instructOpenAlibaba1,401±519,659—
13—longcat-flash-chatOpenMeituan1,399±611,567—
14—qwen3-235b-a22b-no-thinkingOpenAlibaba1,399±539,480—
15—qwen3-235b-a22b-thinking-2507OpenAlibaba1,397±69,374—
16—qwen3-vl-235b-a22b-thinkingOpenAlibaba1,393±78,026—
17—deepseek-v3-0324OpenDeepSeek · 1 more variant1,391±446,909—
18—hunyuan-vision-1.5-thinkingClosedTencent1,390±122,224—
19▲ 2qwen3-coder-480b-a35b-instructOpenAlibaba1,385±523,242—
20—hunyuan-t1-20250711ClosedTencent1,385±94,837—
21▼ 2qwen3-30b-a3b-instruct-2507OpenAlibaba1,382±524,270—
22—hunyuan-turbos-20250416ClosedTencent · 1 more variant1,380±611,144—
23—qwen3-235b-a22bOpenAlibaba1,373±527,211—
24—qwen2.5-maxClosedAlibaba1,372±433,561—
25—glm-4.5-airOpenZ.ai1,370±528,405—
26▲ 1qwen3-next-80b-a3b-thinkingOpenAlibaba1,367±613,904—
27▼ 1minimax-m1OpenMiniMax1,366±436,995—
28—glm-4.5vOpenZ.ai1,351±85,003—
29Newminimax-m2OpenMiniMax1,347±104,217—
30▲ 3ling-flash-2.0OpenAnt Group1,345±77,195—
31▼ 2step-3OpenStepFun1,344±76,673—
32▼ 2qwen3-32bOpenAlibaba1,344±93,943—
33▼ 2qwen-plus-0125ClosedAlibaba1,344±85,861—
34▼ 2glm-4-plus-0111ClosedZhipu · 1 more variant1,342±85,806—
35▼ 1hunyuan-turbo-0110ClosedTencent1,338±112,322—
36▼ 1qwq-32bOpenAlibaba · 1 more variant1,333±426,311—
37▼ 1step-2-16k-exp-202412ClosedStepFun1,331±84,895—
38▼ 1yi-lightningClosed01 AI1,327±527,624—
39▼ 1qwen3-30b-a3bOpenAlibaba1,326±527,526—
40▼ 1hunyuan-large-2025-02-10ClosedTencent1,324±103,760—
41▼ 1deepseek-v2.5-1210OpenDeepSeek · 1 more variant1,321±86,877—
42▲ 1ring-flash-2.0OpenAnt Group1,319±77,317—
43▼ 2step-1o-turbo-202506ClosedStepFun1,319±79,676—
44▼ 2qwen-max-0919OpenAlibaba1,316±616,598—
45▼ 1qwen2.5-plus-1127ClosedAlibaba1,313±610,252—
46▼ 1hunyuan-standard-2025-02-10ClosedTencent1,309±103,920—
47▼ 1qwen2.5-72b-instructOpenAlibaba1,301±439,632—
48▼ 1hunyuan-large-visionClosedTencent1,294±95,603—
49▼ 1glm-4-0520ClosedZhipu AI1,272±79,857—
50▼ 1qwen2.5-coder-32b-instructOpenAlibaba1,268±85,452—

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 “—”.