Skip to content
IFAI

Model rankings

1239 models · 5 boards · 2 sources

Artificial AnalysisIntelligence IndexArenaText arena

Intelligence Index · A composite of several evaluations · Updated May 29, 2026

Smartest
Best open model
Kimi K2.6Kimi53.9Intelligence
Best value
Qwen3.5 0.8B (Reasoning)Alibaba526 pts/$Value
Best from China
Qwen3.7 MaxAlibaba56.6Intelligence
Top three open models
1Kimi K2.6Kimi · #8 overall53.9
2MiMo-V2.5-ProXiaomi · #9 overall53.8
3DeepSeek V4 Pro (Reasoning, Max Effort)DeepSeek · #16 overall51.5
This week's changes
qwen3.7-max-20260517Coding · #3New
xAI Text to SpeechSpeech · #5New
Fun-Realtime-TTS-PreviewSpeech · #6New
RankChangeModelIntelligencePrice $/MSpeed tok/sFirst token
1NewClaude Opus 4.8 (Adaptive Reasoning, Max Effort)ClosedAnthropic61.4$116420 s
2▼ 1GPT-5.5 (xhigh)ClosedOpenAI60.2$116064 s
3▼ 1GPT-5.5 (high)ClosedOpenAI58.9$116320 s
4▼ 1Claude Opus 4.7 (Adaptive Reasoning, Max Effort)ClosedAnthropic57.3$115516 s
5▼ 1Gemini 3.1 Pro PreviewClosedGoogle57.2$4.513322 s
6▼ 1GPT-5.4 (xhigh)ClosedOpenAI56.8$5.684157 s
7▼ 1GPT-5.5 (medium)ClosedOpenAI56.7$11556.2 s
8▼ 1Qwen3.7 MaxClosedAlibaba56.6$3.82012.5 s
9▼ 1Gemini 3.5 Flash (high)ClosedGoogle55.3$3.420617 s
10NewGemini 3.5 Flash (medium)ClosedGoogle54.8$3.419315 s
11▼ 2Kimi K2.6OpenKimi53.9$1.7413.0 s
12▼ 2MiMo-V2.5-ProOpenXiaomi53.8$0.54473.7 s
13▼ 2GPT-5.3 Codex (xhigh)ClosedOpenAI53.6$4.874125 s
14▼ 2Grok 4.3 (high)ClosedxAI53.2$1.617612 s
15▼ 2Claude Opus 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic52.9$114215 s
16▼ 2Muse SparkClosedMeta52.2———
17▼ 2Claude Opus 4.7 (Non-reasoning, High Effort)ClosedAnthropic51.8$11461.4 s
18▼ 2Qwen3.6 Max PreviewClosedAlibaba51.8$2.9353.7 s
19▼ 2Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic51.7$6.661111 s
20▼ 2DeepSeek V4 Pro (Reasoning, Max Effort)OpenDeepSeek51.5$0.54491.8 s
21▼ 2GLM-5.1 (Reasoning)OpenZ AI51.4$2.2581.6 s
22▼ 2GPT-5.2 (xhigh)ClosedOpenAI51.3$4.872122 s
23▼ 2GPT-5.5 (low)ClosedOpenAI50.8$11572.0 s
24▼ 2Qwen3.6 PlusClosedAlibaba50.0$1.1533.1 s
25▼ 2DeepSeek V4 Pro (Reasoning, High Effort)OpenDeepSeek49.8$0.54471.9 s
26▼ 2GLM-5 (Reasoning)OpenZ AI49.8$1.6701.4 s
27▼ 2Claude Opus 4.5 (Reasoning)ClosedAnthropic49.7$114817 s
28▼ 2MiniMax-M2.7OpenMiniMax49.6$0.52732.5 s
29▼ 2Grok 4.20 0309 v2 (Reasoning)ClosedxAI49.3$3.018513 s
30▼ 2MiMo-V2-ProClosedXiaomi49.2$1.5593.0 s
31▼ 2GPT-5.2 Codex (xhigh)ClosedOpenAI49.0$4.810526 s
32▼ 2MiMo-V2.5OpenXiaomi49.0$0.18863.7 s
33▼ 2GPT-5.4 mini (xhigh)ClosedOpenAI48.9$1.71527.5 s
34NewGrok 4.3 (medium)ClosedxAI48.8$1.61637.4 s
35▼ 3Grok 4.20 0309 (Reasoning)ClosedxAI48.5$3.018711 s
36▼ 3Gemini 3 Pro Preview (high)ClosedGoogle48.4$4.513544 s
37▼ 3GPT-5.4 (low)ClosedOpenAI47.9$5.6642.1 s
38▼ 3GPT-5.1 (high)ClosedOpenAI47.7$3.49737 s
39▼ 3Kimi K2.5 (Reasoning)OpenKimi46.8$1.2463.0 s
40▼ 3GLM-5-TurboClosedZ AI46.8———
41▼ 3GPT-5.2 (medium)ClosedOpenAI46.6$4.8——
42▼ 3DeepSeek V4 Flash (Reasoning, Max Effort)OpenDeepSeek46.5$0.181041.2 s
43▼ 3Claude Opus 4.6 (Non-reasoning, High Effort)ClosedAnthropic46.5$11421.8 s
44▼ 3Gemini 3 Flash Preview (Reasoning)ClosedGoogle46.4$1.11737.5 s
45▼ 3DeepSeek V4 Flash (Reasoning, High Effort)OpenDeepSeek46.0$0.18——
46▼ 3Qwen3.6 27B (Reasoning)OpenAlibaba45.8$1.4593.7 s
47▼ 3Qwen3.5 397B A17B (Reasoning)OpenAlibaba45.0$1.4522.5 s
48▼ 3MiMo-V2-Omni-0327ClosedXiaomi44.9$0.801053.5 s
49▼ 3GPT-5 Codex (high)ClosedOpenAI44.6$3.41897.9 s
50▼ 3GPT-5 (high)ClosedOpenAI44.6$3.48876 s

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 “—”.