Atlas

Benchmarks

← All benchmarks

Legal Research Bench Family

Professional Work · 2026-06-23

This row reports the practice-area score for Family tasks in Vals AI Legal Research Bench.

Top models (higher is better)

ModelScore
Claude Opus 540.9
Claude Fable 536.4
Sonnet 4.627.3
Kimi K327.3
GLM-5.222.7
Muse Spark 1.122.7
Opus 4.818.2
Sonnet 518.2
Gemini 3.5 Flash18.2
GPT-5.6 Luna18.2
GPT-5.513.6
GPT-5.6 Sol13.6
GPT-5.6 Terra13.6
Inkling13.6
MiniMax M313.6
DeepSeek-V4-Pro9.1
Gemini 3.6 Flash9.1
GLM-5.19.1
Grok 4.59.1
Qwen3.7-Max9.1
Gemini 3.1 Pro Preview4.5
Gemini 3.5 Flash-Lite4.5
Inkling-Small4.5
GPT-5.4 Mini0.0
Grok 4.30.0
Kimi K2.60.0
Laguna M.10.0
Laguna XS.20.0
Loading Atlas data…