Skip to content
IFAI

Model rankings

906 models · 5 boards · 2 sources

‹
Week of December 29, 2025
February 9, 202641 changesFebruary 2, 20266 changesJanuary 26, 2026177 changesJanuary 19, 202610 changesJanuary 12, 2026116 changesJanuary 5, 20263 changesDecember 29, 202513 changesDecember 22, 20256 changesDecember 15, 20258 changesDecember 8, 202520 changesDecember 1, 202522 changesNovember 24, 202546 changes
›
Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published December 30, 2025

Most preferred
gemini-3-proGoogle1,490Score
Best open model
glm-4.7Z.ai1,439Score
Best from China
ernie-5.0-preview-1203Baidu1,448Score
Most voted
claude-3-opus-20240229Anthropic196,365Votes
Top three by Score
1gemini-3-proGoogle1,490
2gemini-3-flashGoogle1,480
3grok-4.1-thinkingxAI1,477
Top three open models
1glm-4.7Z.ai · #17 overall1,439
2kimi-k2-thinking-turboMoonshot AI · #23 overall1,430
3glm-4.6Z.ai · #25 overall1,425
This week's changes
Speech 2.6 HDSpeech · #2New
Speech 2.6 TurboSpeech · #3New
minimax-m2.1-previewCoding · #5New
Multilingual v2Speech · #7New
GLM-4.7 (Reasoning)Intelligence · #8New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
SortScoreLowest price
RankChangeModelScore95% intervalVotesPrice $/M
1—gemini-3-proClosedGoogle1,490±521,938—
2—gemini-3-flashClosedGoogle1,480±87,020—
3—grok-4.1-thinkingClosedxAI1,477±522,618—
4—claude-opus-4-5-20251101-thinking-32kClosedAnthropic1,470±614,992—
5—claude-opus-4-5-20251101ClosedAnthropic1,467±615,914—
6—grok-4.1ClosedxAI1,466±524,682—
7—gpt-5.1-highClosedOpenAI1,458±519,416—
8—gemini-2.5-proClosedGoogle1,451±382,231—
9—claude-sonnet-4-5-20250929-thinking-32kClosedAnthropic1,450±433,533—
10—claude-opus-4-1-20250805-thinking-16kClosedAnthropic1,448±449,310—
11▲ 1ernie-5.0-preview-1203ClosedBaidu1,448±94,697—
12▼ 1claude-sonnet-4-5-20250929ClosedAnthropic1,446±529,093—
13—gpt-5.2ClosedOpenAI1,443±122,466—
14—gpt-4.5-preview-2025-02-27ClosedOpenAI1,443±614,646—
15—claude-opus-4-1-20250805ClosedAnthropic1,442±362,378—
16—chatgpt-4o-latest-20250326ClosedOpenAI1,441±369,532—
17—gpt-5.2-highClosedOpenAI1,439±96,225—
18Newglm-4.7OpenZ.ai1,439±94,557—
19▼ 1gpt-5.1ClosedOpenAI1,438±520,928—
20▼ 1gpt-5-highClosedOpenAI1,436±532,735—
21▼ 1o3-2025-04-16ClosedOpenAI1,434±461,343—
22▼ 1qwen3-max-previewClosedAlibaba1,434±528,017—
23▼ 1grok-4-1-fast-reasoningClosedxAI1,431±615,352—
24▼ 1kimi-k2-thinking-turboOpenMoonshot AI1,430±521,089—
25▼ 1gpt-5-chatClosedOpenAI1,425±432,013—
26▼ 1glm-4.6OpenZ.ai1,425±429,344—
27▼ 1qwen3-max-2025-09-23ClosedAlibaba1,424±69,212—
28▼ 1deepseek-v3.2-expOpenDeepSeek1,423±611,915—
29▼ 1claude-opus-4-20250514-thinking-16kClosedAnthropic1,423±437,682—
30▼ 1deepseek-v3.2-exp-thinkingOpenDeepSeek1,422±79,181—
31—qwen3-235b-a22b-instruct-2507OpenAlibaba1,421±356,921—
32—grok-4-fast-chatClosedxAI1,421±87,030—
33—deepseek-v3.2OpenDeepSeek1,420±613,952—
34▼ 4deepseek-v3.2-thinkingOpenDeepSeek1,419±711,086—
35—deepseek-r1-0528OpenDeepSeek1,419±619,152—
36▼ 2kimi-k2-0905-previewOpenMoonshot AI1,418±711,942—
37—deepseek-v3.1OpenDeepSeek1,417±615,203—
38—kimi-k2-0711-previewOpenMoonshot AI1,416±528,504—
39—deepseek-v3.1-thinkingOpenDeepSeek1,416±711,939—
40—deepseek-v3.1-terminusOpenDeepSeek1,416±103,733—
41▼ 5mistral-large-3OpenMistral1,415±611,789—
42▼ 1qwen3-vl-235b-a22b-instructOpenAlibaba1,414±710,249—
43▼ 1deepseek-v3.1-terminus-thinkingOpenDeepSeek1,414±103,505—
44▼ 1gpt-4.1-2025-04-14ClosedOpenAI1,412±452,332—
45—mistral-medium-2508ClosedMistral1,412±450,801—
46▼ 2claude-opus-4-20250514ClosedAnthropic1,412±445,430—
47▼ 1grok-3-preview-02-24ClosedxAI1,410±434,035—
48▼ 1grok-4-0709ClosedxAI1,409±442,369—
49▼ 1glm-4.5OpenZ.ai1,409±524,705—
50▼ 1gemini-2.5-flashClosedGoogle1,408±381,715—

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 “—”.