Atlas

Benchmarks

← All benchmarks

Artificial Analysis Intelligence Index v4.3.2

Indexes

Artificial Analysis Intelligence Index v4.3.2, combining AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, and AA-LCR v1.1. This version preserves the source's changed component protocols rather than replacing historical index scores.

Top models (higher is better)

ModelScore
Claude Opus 5.557.6
Claude Fable 5.153.4
GPT-6 Astra52.7
Claude Opus 550.8
Claude Fable 549.6
Muse Spark 1.348.1
GPT-6 Sol47.5
GPT-5.6 Sol47.0
Grok 4.746.4
MiMo-V2.6-Pro46.3
Qwen3.8 Max (0902)45.4
GLM-5.344.8
Grok 4.644.3
Step 5 Preview43.7
Kimi K343.6
GPT-5.6 Terra42.1
GLM-5.3 Flash41.8
Opus 4.841.8
Gemini 3.8 Flash40.9
Opus 4.740.7
Qwen3.8-Max40.2
Qwen3.8 2.4T A95B39.9
Qwen3.8-Flash-Next39.8
Gemini 3.7 Flash39.6
Muse Spark 1.239.6
DeepSeek V4.1 Flash39.5
GPT-5.439.0
Grok 4.538.8
GPT-5.538.4
Sonnet 538.2
GPT-5.6 Luna37.3
GPT-6 Luna37.3
DeepSeek V4 Pro 081336.0
Agnes 3.0 Flash (hosted)35.5
Agnes 2.5 Pro Beta35.2
DeepSeek V4 Flash Vision Exp34.8
DeepSeek-V4-Flash-073134.3
Gemini 3.6 Flash34.0
Muse Spark 1.133.7
GLM-5.233.7
Loading Atlas data…