Skip to content
IFAI

Model rankings

880 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published December 21, 2025

Most preferred
gemini-3-proGoogle1,490Score
Best open model
kimi-k2-thinking-turboMoonshot AI1,428Score
Best from China
ernie-5.0-preview-1203Baidu1,451Score
Most voted
claude-3-opus-20240229Anthropic196,368Votes
Top three by Score
1gemini-3-proGoogle1,490
3gemini-3-flashGoogle1,476
Top three open models
1kimi-k2-thinking-turboMoonshot AI · #22 overall1,428
2glm-4.6Z.ai · #23 overall1,425
3deepseek-v3.2-expDeepSeek · #26 overall1,423
This week's changes
GPT-5.2 (xhigh)Intelligence · #2New
gemini-3-flashCoding · #4New
deepseek-v3.2-thinkingCoding · #12New
Apriel-v1.6-15B-ThinkerIntelligence · #31New
RankChangeModelScore95% intervalVotesPrice $/M
1—gemini-3-proClosedGoogle1,490±619,627—
2—grok-4.1-thinkingClosedxAI · 1 more variant1,477±620,493—
3Newgemini-3-flashClosedGoogle1,476±94,830—
4▼ 1claude-opus-4-5-20251101-thinking-32kClosedAnthropic1,468±613,180—
5▼ 1claude-opus-4-5-20251101ClosedAnthropic1,466±613,981—
6▼ 1gpt-5.1-highClosedOpenAI1,457±617,444—
7▼ 1gemini-2.5-proClosedGoogle1,451±380,262—
8▼ 1claude-sonnet-4-5-20250929-thinking-32kClosedAnthropic1,449±431,612—
9▼ 1claude-opus-4-1-20250805-thinking-16kClosedAnthropic1,448±447,314—
10▼ 1claude-sonnet-4-5-20250929ClosedAnthropic1,446±526,905—
11▼ 1gpt-4.5-preview-2025-02-27ClosedOpenAI1,442±614,644—
12Newgpt-5.2ClosedOpenAI1,442±122,425—
13▼ 2claude-opus-4-1-20250805ClosedAnthropic1,440±460,418—
14▼ 2chatgpt-4o-latest-20250326ClosedOpenAI1,440±367,160—
15Newgpt-5.2-highClosedOpenAI1,440±96,199—
16▼ 3gpt-5.1ClosedOpenAI1,437±618,833—
17▼ 3gpt-5-highClosedOpenAI1,436±532,781—
18▼ 3o3-2025-04-16ClosedOpenAI1,433±461,398—
19▼ 3grok-4-1-fast-reasoningClosedxAI1,430±612,959—
20▼ 3gpt-5-chatClosedOpenAI1,425±432,048—
21▼ 3claude-opus-4-20250514-thinking-16kClosedAnthropic1,423±437,718—
22▼ 3grok-4-fast-chatClosedxAI1,420±87,040—
23▼ 2gpt-4.1-2025-04-14ClosedOpenAI1,412±452,404—
24▼ 4claude-opus-4-20250514ClosedAnthropic1,412±445,477—
25▼ 3grok-3-preview-02-24ClosedxAI1,410±434,039—
26▼ 3grok-4-0709ClosedxAI1,409±442,417—
27▼ 3gemini-2.5-flashClosedGoogle1,408±379,776—
28▼ 3gemini-2.5-flash-preview-09-2025ClosedGoogle1,405±432,981—
29▼ 3claude-haiku-4-5-20251001ClosedAnthropic1,403±429,964$2.0
30▼ 3grok-4-fast-reasoningClosedxAI1,402±518,824—
31▼ 3o1-2024-12-17ClosedOpenAI1,400±428,039—
32▼ 3claude-sonnet-4-20250514-thinking-32kClosedAnthropic1,399±436,072—
33▼ 3gpt-5-mini-highClosedOpenAI1,392±527,348—
34▼ 3o4-mini-2025-04-16ClosedOpenAI1,391±446,706—
35▼ 3mai-1-previewClosedMicrosoft AI1,390±518,119—
36▼ 3claude-sonnet-4-20250514ClosedAnthropic1,389±441,496—
37▼ 3o1-previewClosedOpenAI1,387±531,505—
38▼ 3claude-3-7-sonnet-20250219-thinking-32kClosedAnthropic1,387±439,818—
39▼ 3gpt-4.1-mini-2025-04-14ClosedOpenAI1,380±440,370—
40▼ 3gemini-2.5-flash-lite-preview-09-2025-no-thinkingClosedGoogle1,378±433,010—
41▼ 3gemini-2.5-flash-lite-preview-06-17-thinkingClosedGoogle1,375±433,820—
42▼ 3claude-3-5-sonnet-20241022ClosedAnthropic · 1 more variant1,372±389,733—
43▼ 3claude-3-7-sonnet-20250219ClosedAnthropic1,371±444,475—
44Newamazon-nova-experimental-chat-11-10ClosedAmazon1,367±113,050—
45▼ 4gemma-3-27b-itOpenGoogle1,364±449,180—
46▼ 4o3-mini-highClosedOpenAI1,362±518,735—
47▼ 4grok-3-mini-highClosedxAI1,362±517,505—
48▼ 4gemini-2.0-flash-001ClosedGoogle1,360±445,016—
49▼ 4grok-3-mini-betaClosedxAI1,357±523,702—
50▼ 3gpt-oss-120bOpenOpenAI1,352±431,154$0.26

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 “—”.