Atlas

Benchmarks

← All benchmarks

Harvey HLAB - Criteria Pass Rate - Intellectual Property

Professional Work · 2026-05-06

The Intellectual Property practice-area split of Harvey's Legal Agent Benchmark held-out evaluation criteria pass rate. This reports pooled rubric criteria passed within that split and is distinct from the all-pass task-score split.

Top models (higher is better)

ModelScore
Muse Spark 1.195.4
Grok 4.595.1
Kimi K395.1
Claude Fable 592.6
Qwen3.7-Max91.8
Claude Opus 591.5
GPT-5.6 Luna91.2
Gemini 3.5 Flash90.6
Opus 4.890.5
Kimi K2.690.5
MiniMax M390.5
Sonnet 4.690.1
Gemini 3.6 Flash89.8
Sonnet 588.7
DeepSeek-V4-Pro88.3
GPT-5.586.6
GLM-5.186.2
GPT-5.6 Terra86.2
GPT-5.6 Sol85.5
Inkling84.8
GPT-5.482.3
Qwen3.7-Plus79.8
GLM-5.276.3
Gemini 3.5 Flash-Lite76.0
Grok 4.373.9
Laguna M.165.4
Laguna XS.244.9
Loading Atlas data…