Skip to content
IFAI

Model rankings

984 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published January 29, 2026

Most preferred
gemini-3-proGoogle1,487Score
Best open model
kimi-k2.5-thinkingMoonshot AI1,450Score
Best from China
ernie-5.0-0110Baidu1,453Score
Most voted
claude-3-opus-20240229Anthropic194,904Votes
Top three by Score
1gemini-3-proGoogle1,487
3gemini-3-flashGoogle1,471
Top three open models
1kimi-k2.5-thinkingMoonshot AI · #10 overall1,450
2glm-4.7Z.ai · #17 overall1,440
3kimi-k2-thinking-turboMoonshot AI · #25 overall1,429
This week's changes
grok-imagine-videoVideo · #1New
Inworld TTS 1.5 MaxSpeech · #1New
GPT-5.2 Codex (xhigh)Intelligence · #3New
kimi-k2.5-thinkingCoding · #4New
RankChangeModelScore95% intervalVotesPrice $/M
1—gemini-3-proClosedGoogle1,487±532,205—
2—grok-4.1-thinkingClosedxAI · 1 more variant1,475±432,194—
3—gemini-3-flashClosedGoogle · 1 more variant1,471±522,683—
4—claude-opus-4-5-20251101-thinking-32kClosedAnthropic1,468±523,990—
5—claude-opus-4-5-20251101ClosedAnthropic1,466±528,752—
6—gpt-5.1-highClosedOpenAI1,459±528,296—
7Newernie-5.0-0110ClosedBaidu1,453±78,274—
8—claude-sonnet-4-5-20250929-thinking-32kClosedAnthropic1,450±442,538—
9▲ 1claude-sonnet-4-5-20250929ClosedAnthropic1,450±440,121—
10▼ 3gemini-2.5-proClosedGoogle1,450±391,734—
11—ernie-5.0-preview-1203ClosedBaidu · 1 more variant1,449±79,786—
12▼ 3claude-opus-4-1-20250805-thinking-16kClosedAnthropic1,448±449,975—
13—claude-opus-4-1-20250805ClosedAnthropic1,445±371,681—
14▼ 2gpt-4.5-preview-2025-02-27ClosedOpenAI1,444±614,549—
15▼ 1chatgpt-4o-latest-20250326ClosedOpenAI1,443±378,991—
16▲ 5gpt-5.2ClosedOpenAI1,440±79,387—
17—gpt-5.2-highClosedOpenAI1,440±712,869—
18▼ 2gpt-5.1ClosedOpenAI1,435±530,355—
19▼ 4gpt-5-highClosedOpenAI1,435±532,651—
20▼ 2qwen3-max-previewClosedAlibaba1,434±527,857—
21▼ 2o3-2025-04-16ClosedOpenAI1,433±461,365—
22▼ 2grok-4-1-fast-reasoningClosedxAI1,430±525,264—
23▼ 1gpt-5-chatClosedOpenAI1,426±431,849—
24▼ 1qwen3-max-2025-09-23ClosedAlibaba1,425±69,223—
25▼ 1claude-opus-4-20250514-thinking-16kClosedAnthropic1,424±437,990—
26▼ 1grok-4-fast-chatClosedxAI1,422±86,992—
27—claude-opus-4-20250514ClosedAnthropic1,414±445,572—
28▼ 2gpt-4.1-2025-04-14ClosedOpenAI1,413±452,220—
29▼ 1mistral-medium-2508ClosedMistral · 2 more variants1,412±359,880—
30▼ 1grok-3-preview-02-24ClosedxAI1,411±433,980—
31▼ 1grok-4-0709ClosedxAI1,410±442,174—
32▼ 1gemini-2.5-flashClosedGoogle1,409±390,980—
33▼ 1gemini-2.5-flash-preview-09-2025ClosedGoogle1,405±432,903—
34—grok-4-fast-reasoningClosedxAI1,404±518,646—
35▼ 2claude-haiku-4-5-20251001ClosedAnthropic1,403±441,166$2.0
36▼ 1o1-2024-12-17ClosedOpenAI1,402±427,822—
37▼ 1claude-sonnet-4-20250514-thinking-32kClosedAnthropic1,400±436,212—
38Newamazon-nova-experimental-chat-12-10ClosedAmazon1,395±103,722—
39▼ 1hunyuan-vision-1.5-thinkingClosedTencent1,393±122,222—
40—mai-1-previewClosedMicrosoft AI1,391±518,137—
41▼ 2o4-mini-2025-04-16ClosedOpenAI1,391±446,681—
42▼ 5gpt-5-mini-highClosedOpenAI1,390±527,174—
43▼ 2claude-sonnet-4-20250514ClosedAnthropic1,390±441,643—
44▼ 1claude-3-7-sonnet-20250219-thinking-32kClosedAnthropic1,389±439,890—
45▼ 3o1-previewClosedOpenAI1,388±531,120—
46▼ 2hunyuan-t1-20250711ClosedTencent1,387±94,801—
47▼ 1hunyuan-turbos-20250416ClosedTencent · 1 more variant1,382±611,050—
48▼ 3gpt-4.1-mini-2025-04-14ClosedOpenAI1,382±440,533—
49▼ 2gemini-2.5-flash-lite-preview-09-2025-no-thinkingClosedGoogle1,379±444,231—
50▼ 2gemini-2.5-flash-lite-preview-06-17-thinkingClosedGoogle1,375±433,931—

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 “—”.