Atlas

Benchmarks

← All benchmarks

Legal Research Bench Weighted Pass Health

Professional Work · 2026-06-23

Weighted-pass practice-area score for Health tasks in Vals AI Legal Research Bench.

Top models (higher is better)

ModelScore
GPT-5.6 Sol97.5
Claude Opus 596.8
Gemini 4 Carbon96.8
Grok 4.696.7
Muse Wonder96.5
Ember-196.2
DeepSeek V4.1 Flash 101796.2
Qwen3.8-Max96.2
Gemini 4 Argon96.0
Muse Wonder Full95.8
Grok 4.795.5
GPT-6.1 Sol95.5
Hy4 Preview95.5
GLM-5.395.2
MiMo-V2.6-Pro94.9
DeepSeek V4.1 Flash94.9
GPT-6 Astra94.8
Kimi K394.6
Muse Spark 1.394.6
Claude Haiku 5.594.5
Qwen3.8 27B94.3
GPT-5.6 Luna93.8
Gemini 3.8 Flash93.8
GPT-5.593.6
Grok 4.593.2
GPT-5.6 Terra93.2
GLM-5.3 Flash92.9
Muse Spark 1.292.9
GPT-6.1 Terra92.8
Mistral Large 4 Preview92.4
MiniMax M392.2
Sonnet 592.1
GPT-6 Sol91.7
Gemini 3.5 Flash91.5
GLM-5.291.3
DeepSeek-V4-Pro91.2
GPT-6 Luna91.1
Opus 4.890.8
Muse Spark Open90.8
Gemini 3.7 Flash90.5
Loading Atlas data…