Skip to content
IFAI

Model rankings

1043 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published February 26, 2026

Most preferred
claude-opus-4-6-thinkingAnthropic1,503Score
Best open model
glm-5Z.ai1,455Score
Best from China
dola-seed-2.0-previewByteDance1,470Score
Most voted
claude-3-opus-20240229Anthropic194,904Votes
Top three by Score
1claude-opus-4-6-thinkingAnthropic1,503
3grok-4.20-beta1xAI1,495
Top three open models
1glm-5Z.ai · #13 overall1,455
2qwen3.5-397b-a17bAlibaba · #14 overall1,454
3kimi-k2.5-thinkingMoonshot AI · #16 overall1,452
This week's changes
Kling 3.0 Omni ProVideo · #2New
Kling 3.0 Omni StandardVideo · #10New
qwen3.5-397b-a17bCoding · #14New
Recraft V4 ProImage · #34New
RankChangeModelScore95% intervalVotesPrice $/M
1—claude-opus-4-6-thinkingClosedAnthropic1,503±86,583—
2—claude-opus-4-6ClosedAnthropic1,503±77,454—
3—gemini-3.1-pro-previewClosedGoogle1,500±94,052—
4—grok-4.20-beta1ClosedxAI1,495±103,818—
5—gemini-3-proClosedGoogle1,486±438,248—
6—gpt-5.2-chat-latest-20260210ClosedOpenAI1,481±103,605—
7▲ 1gemini-3-flashClosedGoogle1,473±529,334—
8▲ 1grok-4.1-thinkingClosedxAI1,473±437,474—
9▲ 1claude-opus-4-5-20251101-thinking-32kClosedAnthropic1,471±430,541—
10▼ 3dola-seed-2.0-previewClosedByteDance1,470±94,620—
11—claude-opus-4-5-20251101ClosedAnthropic1,467±435,476—
12—grok-4.1ClosedxAI1,462±441,700—
13—gemini-3-flash (thinking-minimal)ClosedGoogle1,461±520,672—
14▲ 1claude-sonnet-4-6ClosedAnthropic1,458±103,470—
15▼ 1gpt-5.1-highClosedOpenAI1,457±434,379—
16—glm-5OpenZ.ai1,455±76,466—
17▲ 5qwen3.5-397b-a17bOpenAlibaba1,454±84,958—
18▼ 1ernie-5.0-0110ClosedBaidu1,453±613,833—
19▼ 1kimi-k2.5-thinkingOpenMoonshot AI1,452±611,075—
20—claude-sonnet-4-5-20250929-thinking-32kClosedAnthropic1,450±348,912—
21▼ 2claude-sonnet-4-5-20250929ClosedAnthropic1,450±446,720—
22▼ 1gemini-2.5-proClosedGoogle1,449±397,296—
23—ernie-5.0-preview-1203ClosedBaidu1,449±79,725—
24—claude-opus-4-1-20250805-thinking-16kClosedAnthropic1,449±449,597—
25—claude-opus-4-1-20250805ClosedAnthropic1,446±377,218—
26—gpt-4.5-preview-2025-02-27ClosedOpenAI1,444±614,549—
27—chatgpt-4o-latest-20250326ClosedOpenAI1,443±382,938—
28▲ 1gpt-5.2-highClosedOpenAI1,441±519,253—
29▼ 1glm-4.7OpenZ.ai1,441±611,937—
30▲ 1gpt-5.1ClosedOpenAI1,437±436,738—
31▼ 1gpt-5.2ClosedOpenAI1,437±516,113—
32▲ 2qwen3-max-previewClosedAlibaba1,434±527,642—
33▼ 1kimi-k2.5-instantOpenMoonshot AI1,434±77,125—
34▼ 1gpt-5-highClosedOpenAI1,434±532,346—
35—o3-2025-04-16ClosedOpenAI1,432±460,957—
36—grok-4-1-fast-reasoningClosedxAI1,431±431,128—
37—kimi-k2-thinking-turboOpenMoonshot AI1,429±436,099—
38—gpt-5-chatClosedOpenAI1,426±431,603—
39—glm-4.6OpenZ.ai1,425±435,128—
40—qwen3-max-2025-09-23ClosedAlibaba1,425±69,170—
41—claude-opus-4-20250514-thinking-16kClosedAnthropic1,424±437,722—
42▲ 1deepseek-v3.2-expOpenDeepSeek1,424±611,680—
43▼ 1deepseek-v3.2-exp-thinkingOpenDeepSeek1,423±78,944—
44Newqwen3.5-122b-a10bOpenAlibaba1,423±28366—
45▼ 1qwen3-235b-a22b-instruct-2507OpenAlibaba1,422±371,551—
46▼ 1grok-4-fast-chatClosedxAI1,422±86,962—
47▼ 1deepseek-v3.2-thinkingOpenDeepSeek1,420±525,692—
48▼ 1deepseek-v3.2OpenDeepSeek1,420±430,709—
49▼ 1deepseek-r1-0528OpenDeepSeek1,419±619,177—
50▼ 1ernie-5.0-preview-1022ClosedBaidu1,419±94,561—

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 “—”.