Every AI leaderboard

Leaderboards28Models ranked992

28 leaderboards · 992 models ranked

Epoch Capabilities Index what it measures, how it is scored and who leads it

Composite indexes · Epoch AI · ECI

Leaderboard

Scores from Epoch Capabilities Index, published Nov 11, 2026

RankModelECI
1Claude Fable 5.2Anthropic · as “Claude Fable 5.2”174.2
2GPT-6 AstraOpenAI · as “GPT-6 Astra”169.2
3GPT-6.1 SolOpenAI · as “GPT-6.1 Sol”168.9
4Claude Opus 5.5Anthropic · as “Claude Opus 5.5”168.7
5GPT-6.1 TerraOpenAI · as “GPT-6.1 Terra”168.1
6Claude Fable 5.1Anthropic · as “Claude Fable 5.1”167.0
7Gemini 4 ArgonGoogle · as “Gemini 4 Argon”166.5
8Claude Sonnet 5.5Anthropic · as “Claude Sonnet 5.5”166.3
9Claude Opus 5Anthropic · as “Claude Opus 5”164.5
10Claude Fable 5Anthropic · as “Claude Fable 5”164.3
11GPT-6 SolOpenAI · as “GPT-6 Sol”164.0
12GPT-5.6 SolOpenAI · as “GPT-5.6 Sol”163.0
13GPT-5.5 ProOpenAI · as “GPT-5.5 Pro”162.8
14Kimi K3.1 ProMoonshot AI · as “Kimi K3.1 Pro”162.0
15Muse WonderMeta · as “Muse Wonder”162.0
16GPT-5.6 TerraOpenAI · as “GPT-5.6 Terra”160.9
17Claude Haiku 5.5Anthropic · as “Claude Haiku 5.5”160.3
18GPT-5.4 ProOpenAI · as “GPT-5.4 Pro”160.2
19GPT-5.5OpenAI · as “GPT-5.5”160.1
20Claude Opus 4.8Anthropic · as “Claude Opus 4.8”159.9
21DeepSeek V4.1 FlashDeepSeek · as “DeepSeek V4.1 Flash 1017”159.3
22Gemini 3.8 FlashGoogle · as “Gemini 3.8 Flash”158.7
23Kimi K3Moonshot AI · as “Kimi K3”158.5
24Gemini 3.7 FlashGoogle · as “Gemini 3.7 Flash”158.4
25Muse Spark 1.3Meta · as “Muse Spark 1.3”158.0

About this benchmark

Task
One capability score stitched together from dozens of benchmarks.
Dataset
Scores from Epoch AI's benchmarking hub and published results.
Method
Item response theory places models and benchmarks on one scale.
Metric
ECI
Organization
Epoch AI
Official leaderboard
epoch.ai/eci

Sources: each benchmark's paper (arXiv) and its official site or leaderboard. Scores are the boards captured here every day, exactly as published.