Skip to content
IFAI

Model rankings

1190 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published May 1, 2026

Most preferred
claude-opus-4-7-thinkingAnthropic1,503Score
Best open model
glm-5.1Z.ai1,471Score
Best from China
ernie-5.1-previewBaidu1,475Score
Most voted
claude-3-opus-20240229Anthropic194,909Votes
Top three by Score
1claude-opus-4-7-thinkingAnthropic1,503
2claude-opus-4-6-thinkingAnthropic1,502
Top three open models
1glm-5.1Z.ai · #16 overall1,471
2mimo-v2.5-proXiaomi · #20 overall1,463
3deepseek-v4-proDeepSeek · #22 overall1,463
This week's changes
Grok 4.3Intelligence · #8New
mimo-v2.5-proCoding · #9New
gpt-5.5 (codex-harness)Coding · #15New
RankChangeModelScore95% intervalVotesPrice $/M
1—claude-opus-4-7-thinkingClosedAnthropic · 1 more variant1,503±77,615—
2—claude-opus-4-6-thinkingClosedAnthropic · 1 more variant1,502±522,385—
3—gemini-3.1-pro-previewClosedGoogle1,493±528,096—
4—muse-sparkClosedMeta1,491±69,414—
5Newgpt-5.5-highClosedOpenAI1,488±85,121—
6▼ 1gemini-3-proClosedGoogle1,486±441,369—
7▼ 1grok-4.20-beta1ClosedxAI1,480±517,413—
8—grok-4.20-beta-0309-reasoningClosedxAI1,477±516,204—
9▼ 2gpt-5.4-highClosedOpenAI1,477±515,853—
10▼ 1gpt-5.2-chat-latest-20260210ClosedOpenAI1,476±522,311—
11Newgpt-5.5ClosedOpenAI1,475±85,302—
12▼ 2grok-4.20-multi-agent-beta-0309ClosedxAI1,475±516,458—
13Newernie-5.1-previewClosedBaidu1,475±84,738—
14▼ 3gemini-3-flashClosedGoogle · 1 more variant1,474±430,790—
15▼ 3claude-opus-4-5-20251101-thinking-32kClosedAnthropic1,473±437,158—
16▼ 3glm-5.1OpenZ.ai1,471±611,071—
17▼ 3grok-4.1-thinkingClosedxAI · 1 more variant1,468±353,942—
18▼ 3claude-opus-4-5-20251101ClosedAnthropic1,468±353,597—
19▼ 3gpt-5.4ClosedOpenAI1,466±516,567—
20Newmimo-v2.5-proOpenXiaomi1,463±85,037—
21▼ 2claude-sonnet-4-6ClosedAnthropic1,463±515,794—
22▼ 4deepseek-v4-proOpenDeepSeek · 1 more variant1,463±94,176—
23▼ 6qwen3.5-max-previewClosedAlibaba1,463±613,565—
24▼ 3kimi-k2.6OpenMoonshot AI1,460±76,168—
25▼ 5dola-seed-2.0-proClosed1,459±525,273—
26Newqwen3.6-max-previewClosedAlibaba1,457±93,943—
27▼ 4glm-5OpenZ.ai1,456±519,980—
28▼ 6gpt-5.4-mini-highClosedOpenAI1,456±613,541—
29Newgrok-4.3ClosedxAI1,456±112,505—
30▼ 6gpt-5.1-highClosedOpenAI1,455±440,879—
31▼ 6claude-sonnet-4-5-20250929-thinking-32kClosedAnthropic1,454±365,936—
32▼ 6claude-sonnet-4-5-20250929ClosedAnthropic1,453±363,899—
33▼ 6gemma-4-31bOpenGoogle1,451±85,810$0.000
34▼ 6ernie-5.0-0110ClosedBaidu1,451±427,672—
35▼ 4kimi-k2.5-thinkingOpenMoonshot AI1,449±426,123—
36▼ 6ernie-5.0-preview-1203ClosedBaidu · 1 more variant1,449±79,770—
37▼ 3mimo-v2-proClosedXiaomi1,449±614,085—
38▼ 6claude-opus-4-1-20250805-thinking-16kClosedAnthropic1,449±349,853—
39▼ 10gpt-5.3-chat-latestClosedOpenAI1,448±521,109—
40▼ 7gemini-2.5-proClosedGoogle1,448±3113,545—
41▼ 5qwen3.6-plusClosedAlibaba1,447±77,538—
42▼ 5qwen3.5-397b-a17bOpenAlibaba1,447±521,382—
43▼ 8claude-opus-4-1-20250805ClosedAnthropic1,447±377,437—
44▼ 6gpt-4.5-preview-2025-02-27ClosedOpenAI1,444±614,547—
45▼ 6chatgpt-4o-latest-20250326ClosedOpenAI1,443±382,545—
46▼ 6glm-4.7OpenZ.ai1,443±612,143—
47▼ 6gpt-5.2-highClosedOpenAI1,439±436,731—
48▼ 5gemini-3.1-flash-lite-previewClosedGoogle1,439±522,387$0.56
49▼ 4gpt-5.1ClosedOpenAI1,439±443,516—
50▼ 8deepseek-v4-flash-thinkingOpenDeepSeek · 1 more variant1,439±93,603—

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 “—”.