Atlas

Benchmarks

← All benchmarks

Legal Research Bench Constitutional/Civil Rights

Professional Work · 2026-06-23

This row reports the practice-area score for Constitutional/Civil Rights tasks in Vals AI Legal Research Bench.

Top models (higher is better)

ModelScore
Gemini 4 Carbon55.3
Muse Spark 1.350.0
Gemini 4 Argon47.4
MiMo-V2.6-Pro47.4
Muse Wonder Full47.4
Claude Haiku 5.544.7
Opus 4.744.7
Claude Opus 544.7
Sonnet 544.7
GLM-5.3 Flash44.7
Muse Spark Open44.7
Qwen3.8 Max (0902)44.7
DeepSeek V4.1 Flash 101742.1
Ember-142.1
Gemini 3.8 Flash42.1
Grok 4.742.1
Opus 4.839.5
DeepSeek V4 Pro 081339.5
GLM-5.339.5
Hy4 Preview39.5
Muse Spark 1.239.5
Muse Wonder39.5
DeepSeek V4.1 Flash36.8
Grok 4.636.8
Kimi K336.8
Sonnet 4.634.2
GLM-5.134.2
GPT-5.6 Luna34.2
GPT-5.6 Sol34.2
GPT-5.6 Terra34.2
Mercury 334.2
MiMo-V2.6-Flash34.2
MiniMax M334.2
Mistral Large 4 Preview34.2
Qwen3.8-Max34.2
Step 5 Preview34.2
Beam 1.0 501B A23B31.6
DeepSeek-V4-Flash-073131.6
Gemini 3.7 Flash31.6
GPT-6.1 Sol31.6
Loading Atlas data…