Skip to content
IFAI

Model rankings

896 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published November 9, 2026

Most preferred
Gemini 4 Carbon (High)Google1,526Score
Best open model
Kimi K3 (Max)Moonshot AI1,487Score
Best from China
Kimi K3 (Max)Moonshot AI1,487Score
Most voted
Claude 3 Opus (20240229)Anthropic194,909Votes
Top three by Score
Top three open models
1Kimi K3 (Max)Moonshot AI · #16 overall1,487
2Kimi K3.1 Pro (Max)Moonshot AI · #21 overall1,483
3GLM-5.3 (Max)Z.ai · #23 overall1,483
This week's changes
Simba 3.5Speech · #5New
Bach 1.0 ProVideo · #7New
RankChangeModelScore95% intervalVotesPrice $/M
1—Gemini 4 Carbon (High)ClosedGoogle1,526±104,282$8.0
2—Gemini 4 Argon (High)ClosedGoogle1,522±519,787$8.0
3—Muse Wonder (Max)ClosedMeta1,511±612,032$8.0
4—Claude Opus 5.5 (High)ClosedAnthropic1,506±615,249$8.0
5—Claude Fable 5 (High)ClosedAnthropic1,504±445,524—
6▲ 1Claude Opus 4.6 (High)ClosedAnthropic · 1 more variant1,503±384,424—
7▼ 1Claude Fable 5.1 (Max)ClosedAnthropic1,503±620,087$20
8—Claude Opus 4.7 (High)ClosedAnthropic · 1 more variant1,501±464,308—
9—Gemini 3.8 Flash (High)ClosedGoogle1,497±461,143$1.5
10—Muse Spark 1.3 (Max)ClosedMeta1,496±526,111$2.0
11—Muse Spark 1.2 (xHigh)ClosedMeta1,493±615,004—
12—Muse Spark 1.1ClosedMeta1,491±445,562—
13—Claude Opus 5 (High)ClosedAnthropic · 1 more variant1,490±473,697$10
14—Gemini 3.7 Flash (High)ClosedGoogle1,489±533,300—
15—Muse SparkClosedMeta1,489±614,009—
16—Gemini 3.1 Pro PreviewClosedGoogle1,487±3132,296—
17—Gemini 3 ProClosedGoogle1,486±441,829—
18—GPT-5.6 Sol (xHigh)ClosedOpenAI1,485±443,994—
19—Gemini 3.6 Flash (High)ClosedGoogle1,483±435,405—
20▲ 1GPT-6.1 Sol (Max)ClosedOpenAI1,482±612,952$4.0
21▼ 1Claude Opus 4.8 (High)ClosedAnthropic · 1 more variant1,482±463,128—
22—GPT-5.5 (High)ClosedOpenAI · 1 more variant1,481±468,359—
23—Gemini 3.5 Flash (High)ClosedGoogle · 1 more variant1,478±447,642—
24—GPT-5.2 Chat (20260210)ClosedOpenAI1,476±435,708—
25—GPT-5.4 (High)ClosedOpenAI · 1 more variant1,475±464,103—
26—Grok 4.20 Beta 1ClosedxAI1,475±527,690—
27—GPT-6 Astra (Max)ClosedOpenAI1,474±618,315$20
28—Claude Opus 4.5 (High, 32K, 20251101)ClosedAnthropic · 1 more variant1,474±437,253—
29—GPT-5.5 InstantClosedOpenAI1,473±526,865—
30—Gemini 3 FlashClosedGoogle · 1 more variant1,473±431,133—
31▲ 2Claude Sonnet 5.5 (xHigh)ClosedAnthropic1,472±613,821$4.0
32—Grok 4.20 Beta (Reasoning, 0309)ClosedxAI1,472±465,713—
33▼ 2Claude Sonnet 4.6ClosedAnthropic1,472±470,017—
34—Grok 4.20 Multi-Agent Beta (0309)ClosedxAI1,471±464,252—
35—GPT-5.6 Terra (xHigh)ClosedOpenAI1,468±448,825$4.5
36—Grok 4.5ClosedxAI1,466±439,155—
37—Grok 4.1 (Thinking)ClosedxAI · 1 more variant1,465±366,074—
38—Claude Sonnet 5 (High)ClosedAnthropic1,461±453,013—
39—GPT-6 Sol (Max)ClosedOpenAI1,457±619,890$4.0
40—Grok 4.6 (High)ClosedxAI1,456±534,616$3.0
41—Claude Sonnet 4.5 (High, 32K, 20250929)ClosedAnthropic · 1 more variant1,456±383,387—
42—GPT-5.1 (High)ClosedOpenAI · 1 more variant1,455±441,405—
43—Gemini 3.5 Flash-LiteClosedGoogle1,455±535,114$0.85
44—GPT-5.6 Luna (xHigh)ClosedOpenAI1,453±440,641$0.45
45—GPT-5.3 ChatClosedOpenAI1,450±434,076—
46—Claude Opus 4.1 (Thinking 16K, 20250805)ClosedAnthropic · 1 more variant1,450±349,374—
47—GPT-5.4 mini (High)ClosedOpenAI1,447±462,795—
48—Gemini 2.5 ProClosedGoogle1,446±2124,789—
49—GPT-4.5 Preview (2025-02-27)ClosedOpenAI1,445±614,547—
50—ChatGPT-4o Latest (20250326)ClosedOpenAI1,443±381,663—

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 “—”.