Skip to content
IFAI

Model rankings

1319 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published July 2, 2026

Most preferred
claude-fable-5Anthropic1,509Score
Best open model
glm-5.1Z.ai1,472Score
Best from China
qwen3.7-max-previewAlibaba1,475Score
Most voted
claude-3-opus-20240229Anthropic194,909Votes
Top three by Score
1claude-fable-5Anthropic1,509
2claude-opus-4-6-thinkingAnthropic1,504
3claude-opus-4-7-thinkingAnthropic1,502
Top three open models
1glm-5.1Z.ai · #19 overall1,472
2glm-5.2 (max)Z.ai · #23 overall1,469
3mimo-v2.5-proXiaomi · #26 overall1,466
This week's changes
Reve 2.0Image · #2New
claude-sonnet-5-thinkingCoding · #5New
Fish Audio S2.1 ProSpeech · #15New
OpenAudio S1 MiniSpeech · #50New
RankChangeModelScore95% intervalVotesPrice $/M
1—claude-fable-5ClosedAnthropic1,509±94,350—
2—claude-opus-4-6-thinkingClosedAnthropic · 1 more variant1,504±455,102—
3—claude-opus-4-7-thinkingClosedAnthropic · 1 more variant1,502±441,868—
4—muse-sparkClosedMeta1,487±613,591—
5—gemini-3.1-pro-previewClosedGoogle1,486±473,099—
6—gemini-3-proClosedGoogle1,486±441,306—
7—claude-opus-4-8-thinkingClosedAnthropic · 1 more variant1,484±522,340—
8—gpt-5.5-highClosedOpenAI1,481±537,260—
9▲ 1gemini-3.5-flashClosedGoogle1,479±615,261—
10▼ 1gpt-5.4-highClosedOpenAI1,478±450,378—
11—gpt-5.2-chat-latest-20260210ClosedOpenAI1,476±434,518—
12▲ 1qwen3.7-max-previewClosedAlibaba1,475±103,727—
13▲ 1gpt-5.5ClosedOpenAI1,475±538,470—
14▼ 2grok-4.20-beta-0309-reasoningClosedxAI1,475±451,724—
15—grok-4.20-beta1ClosedxAI1,474±526,920—
16—gemini-3-flashClosedGoogle · 1 more variant1,473±430,711—
17—claude-opus-4-5-20251101-thinking-32kClosedAnthropic1,473±437,085—
18—gpt-5.5-instantClosedOpenAI1,473±526,172—
19—claude-sonnet-4-6ClosedAnthropic1,472±448,706—
20—grok-4.20-multi-agent-beta-0309ClosedxAI1,470±450,647—
21—claude-opus-4-5-20251101ClosedAnthropic1,469±371,135—
22—ernie-5.1ClosedBaidu1,468±532,742—
23—gpt-5.4ClosedOpenAI1,467±453,056—
24—grok-4.1-thinkingClosedxAI · 1 more variant1,466±365,565—
25—qwen3.5-max-previewClosedAlibaba1,465±521,531—
26Newclaude-sonnet-5-thinkingClosedAnthropic1,464±94,385—
27▼ 1qwen3.7-plusClosedAlibaba1,463±614,900—
28▼ 1qwen3.6-max-previewClosedAlibaba1,460±85,215—
29▼ 1claude-sonnet-4-5-20250929-thinking-32kClosedAnthropic1,455±382,456—
30—claude-sonnet-4-5-20250929ClosedAnthropic1,455±380,895—
31▼ 2dola-seed-2.0-proClosedByteDance1,455±459,361—
32▼ 1gpt-5.1-highClosedOpenAI1,455±440,823—
33▲ 1gpt-5.4-mini-highClosedOpenAI1,449±448,963—
34▼ 2claude-opus-4-1-20250805-thinking-16kClosedAnthropic1,449±349,773—
35▼ 2ernie-5.0-preview-1203ClosedBaidu · 1 more variant1,449±79,746—
36▼ 1gpt-5.3-chat-latestClosedOpenAI1,449±433,084—
37▼ 1mimo-v2-proClosedXiaomi1,448±524,557—
38▲ 1claude-opus-4-1-20250805ClosedAnthropic1,447±377,292—
39▼ 1ernie-5.0-0110ClosedBaidu1,447±435,291—
40—gemini-2.5-proClosedGoogle1,446±3124,527—
41—gpt-4.5-preview-2025-02-27ClosedOpenAI1,445±614,547—
42▲ 2grok-4.3ClosedxAI1,444±537,636—
43▼ 1qwen3.6-plusClosedAlibaba1,443±536,564—
44▼ 1chatgpt-4o-latest-20250326ClosedOpenAI1,443±382,424—
45—gpt-5.1ClosedOpenAI1,439±443,455—
46—gpt-5.2-highClosedOpenAI1,437±448,052—
47—longcat-flash-chat-2602-expClosedMeituan1,436±528,162—
48—gpt-5.2ClosedOpenAI1,435±368,870—
49—qwen3-max-previewClosedAlibaba1,435±427,707—
50—gpt-5-highClosedOpenAI1,434±531,915—

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 “—”.