The Aggregate — unified LLM rankings, daily

Unified rankings for 1776 AI models across 5206 public benchmarks, aggregated from public leaderboards and refreshed daily. Rankings use an Item Response Theory (IRT) fit that maps every model onto one ELO scale.

Top models by unified ELO

RankModelProviderUnified ELOBenchmarks
1Human ExpertHuman2500 ± 4859
2GPT-5.5 Pro (xHigh)OpenAI2268 ± 5712
3Claude Mythos 5Anthropic2250 ± 3157
4Gemini 3 Deep ThinkGoogle2186 ± 7118
5GPT-5.6 Sol (Max)OpenAI2175 ± 4838
6Claude Fable 5 (Max)Anthropic2173 ± 3727
7Claude Opus 4.6 (Medium)Anthropic2129 ± 1317
8Claude Fable 5 (High)Anthropic2104 ± 4118
9Claude Mythos PreviewAnthropic2103 ± 2179
10GPT-5.6 Sol (xHigh)OpenAI2096 ± 3942
11GPT-5.5 ProOpenAI2092 ± 6329
12Claude Fable 5Anthropic2088 ± 25140
13GPT-5.5 (xHigh)OpenAI2057 ± 17109
14GPT-5.6 Terra (Max)OpenAI2051 ± 3931
15GPT-5.6 Sol (High)OpenAI2041 ± 3328
16Kimi K3 (Max)Moonshot2034 ± 5211
17GPT-5.4 Pro (xHigh)OpenAI2019 ± 2763
18GPT-5.6 Sol (Medium)OpenAI2016 ± 3123
19GPT-5.2 (xHigh)OpenAI2012 ± 2685
20GPT-5.5 (High)OpenAI2011 ± 2666

Latest leaderboard changes

Explore: all models · benchmark catalog · capability trends · daily changes

Interactive version: theaggregate.ai/ · How the rankings work · Data refreshed daily, snapshot 2026-07-22.