Atlas

Benchmarks

← All benchmarks

HLAB-Vals-1.1 - Criteria Pass Rate - Employment Labor

Professional Work · 2026-10-19

Criteria pass rate (judge-averaged, Harvey's definition) on the Employment Labor practice-area tasks of HLAB-Vals-1.1, distinct from that area's task pass rate.

Top models (higher is better)

ModelScore
Muse Wonder Full93.4
Muse Wonder93.1
Gemini 4 Argon92.9
Muse Spark Open92.5
DeepSeek V4.1 Flash 101790.6
Grok 4.790.1
DeepSeek V4 Pro 081389.3
GPT-6 Astra89.1
Mistral Large 4 Preview89.1
GLM-5.388.8
Kimi K388.8
Gemini 3.8 Flash88.6
MiMo-V2.6-Flash88.6
MiMo-V2.6-Pro88.2
GPT-6.1 Sol88.0
Claude Haiku 5.587.3
Hy4 Preview86.3
Claude Opus 5.585.2
GLM-5.3 Flash84.6
Mercury 384.3
GPT-6 Luna83.7
Beam 1.0 501B A23B83.5
MiniMax M3.1 Flash82.2
Claude Sonnet 5.580.2
Inkling79.4
Inkling-Small77.9
Loading Atlas data…