Atlas

Benchmarks

← All benchmarks

Artificial Analysis Engineering Index

Indexes

Artificial Analysis' capability index for engineering work: 30% AA-Omniscience on its science, engineering and mathematics domain, 20% the mean of the CritPt v2 hard subset and Terminal-Bench-Science 0.1, 20% HLE-Diamond, 15% the mean of Terminal-Bench 4.0 and FrontierCode 2.0 and 15% AA-Briefcase v1.1. It reweights evaluations Atlas already has, so it is a diagnostic. Published only at /indexes/engineering.

Benchmark source: Artificialanalysis

Top models (higher is better)

ModelScoreSource
Gemini 4 Carbon65.3Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
GPT-6 Astra61.8Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
GPT-6.1 Sol59.4Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Claude Sonnet 5.558.5Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Muse Wonder Full55.3Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
GPT-6.1 Terra54.7Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Kimi K3.1 Pro50.9Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Muse Wonder49.6Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Claude Haiku 5.542.5Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Grok 4.741.8Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Qwen3.8 Max (0902)39.3Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Gemini 3.8 Flash38.8Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
DeepSeek V4.1 Flash 101737.6Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
GLM-5.337.6Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Step 5 Preview37.0Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Ling 3.1 Flash 101935.5Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Muse Spark Open35.1Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
MiniMax M3.1 Flash34.3Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
MiMo-V2.6-Pro34.3Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Mistral Large 433.4Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
GPT-6 Luna33.3Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Qwen3.8 27B28.1Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
K2 Horizon 375B A23B26.6Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Beam 1.0 501B A23B25.2Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Nemotron 3.5 Super VL 120B A12B (precision unspecified)23.5Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Inkling22.0Artificial Analysis Engineering Index (snapshot 2026-11-03) · Artificial Analysis
Loading Atlas data…