Skip to content
IFAI

Model rankings

326 models · 2 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published May 11, 2025

Most preferred
o3-2025-04-16OpenAI1,375Score
Best open model
deepseek-v3-0324DeepSeek1,323Score
Best from China
deepseek-v3-0324DeepSeek1,323Score
Most voted
claude-3-opus-20240229Anthropic202,651Votes
Top three by Score
1o3-2025-04-16OpenAI1,375
Top three open models
1deepseek-v3-0324DeepSeek · #7 overall1,323
2deepseek-r1DeepSeek · #9 overall1,316
3qwen3-235b-a22bAlibaba · #18 overall1,286
This week's changes
Qwen3 235B A22B (Reasoning)Intelligence · #6New
Qwen3 32B (Reasoning)Intelligence · #11New
Qwen3 14B (Reasoning)Intelligence · #14New
Qwen3 30B A3B (Reasoning)Intelligence · #15New
Nova PremierIntelligence · #37New
RankChangeModelScore95% intervalVotesPrice $/M
1—deepseek-v3-0324OpenDeepSeek · 1 more variant1,323±68,753—
2—deepseek-r1OpenDeepSeek1,316±418,493—
3▲ 1hunyuan-turbos-20250416ClosedTencent · 1 more variant1,295±103,699—
4▼ 1qwen2.5-maxClosedAlibaba1,286±423,180—
5Newqwen3-235b-a22bOpenAlibaba1,286±93,611—
6▼ 1qwen-plus-0125ClosedAlibaba1,263±76,058—
7▼ 1hunyuan-turbo-0110ClosedTencent1,259±112,512—
8—glm-4-plus-0111ClosedZhipu · 1 more variant1,257±76,024—
9▼ 2qwq-32bOpenAlibaba · 1 more variant1,256±59,946—
10▼ 1step-2-16k-exp-202412ClosedStepFun1,252±75,125—
11▼ 1yi-lightningClosed01 AI1,247±328,968—
12▼ 1hunyuan-large-2025-02-10ClosedTencent1,242±93,858—
13▼ 1deepseek-v2.5-1210OpenDeepSeek · 1 more variant1,241±57,242—
14—qwen2.5-plus-1127ClosedAlibaba1,232±410,711—
15▼ 2qwen-max-0919OpenAlibaba1,232±417,432—
16▼ 1hunyuan-standard-2025-02-10ClosedTencent1,230±94,015—
17▼ 1qwen2.5-72b-instructOpenAlibaba1,223±241,517—
18▼ 1qwen2.5-coder-32b-instructOpenAlibaba1,190±75,733—
19▼ 1glm-4-0520ClosedZhipu AI1,189±510,224—
20▼ 1deepseek-coder-v2OpenDeepSeek1,182±415,756—
21▼ 1qwen2-72b-instructOpenAlibaba1,181±338,870—
22▼ 1qwen1.5-110b-chatOpenAlibaba1,155±427,438—
23▼ 1hunyuan-standard-256kClosedTencent1,153±112,900—
24▼ 1qwen1.5-72b-chatOpenAlibaba1,148±340,665—
25▼ 1yi-1.5-34b-chatOpen01 AI1,134±325,136—
26▼ 1qwen1.5-32b-chatOpenAlibaba1,126±422,762—
27▼ 1qwen1.5-14b-chatOpenAlibaba1,107±418,685—
28▼ 1yi-34b-chatOpen01 AI1,101±415,918—
29▼ 1deepseek-llm-67b-chatOpenDeepSeek1,091±84,988—
30▼ 1qwen1.5-7b-chatOpenAlibaba1,065±84,873—
31▼ 1qwen-14b-chatOpenAlibaba1,056±95,067—
32▼ 1qwen1.5-4b-chatOpenAlibaba1,014±77,819—

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 “—”.