Atlas

Benchmarks

← All benchmarks

HLAB-Vals-1.1 - White Collar Defense Investigations

Professional Work · 2026-10-19

Mean of two judges' task pass rates on the White Collar Defense Investigations practice-area tasks of HLAB-Vals-1.1.

Top models (higher is better)

ModelScore
Muse Wonder20.0
Beam 1.0 501B A23B0.0
Claude Haiku 5.50.0
Claude Opus 5.50.0
Claude Sonnet 5.50.0
DeepSeek V4.1 Flash 10170.0
DeepSeek V4 Pro 08130.0
Gemini 3.8 Flash0.0
Gemini 4 Argon0.0
GLM-5.3 Flash0.0
GLM-5.30.0
GPT-6.1 Sol0.0
GPT-6 Astra0.0
GPT-6 Luna0.0
Grok 4.70.0
Hy4 Preview0.0
Inkling0.0
Inkling-Small0.0
Kimi K30.0
Mercury 30.0
MiMo-V2.6-Flash0.0
MiMo-V2.6-Pro0.0
MiniMax M3.1 Flash0.0
Mistral Large 4 Preview0.0
Muse Spark Open0.0
Muse Wonder Full0.0
Loading Atlas data…