Harvey HLAB - Criteria Pass Rate - Energy Natural Resources
Professional Work · 2026-05-06
The Energy Natural Resources 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)
| Model | Score |
|---|---|
| Opus 4.8 | 97.0 |
| Kimi K3 | 96.7 |
| Muse Spark 1.1 | 96.4 |
| Claude Fable 5 | 96.0 |
| Claude Opus 5 | 95.0 |
| Sonnet 4.6 | 95.0 |
| Grok 4.5 | 95.0 |
| GPT-5.5 | 93.7 |
| DeepSeek-V4-Pro | 93.4 |
| MiniMax M3 | 93.4 |
| GPT-5.6 Sol | 92.7 |
| GPT-5.6 Luna | 91.7 |
| Sonnet 5 | 91.4 |
| Qwen3.7-Max | 88.4 |
| Gemini 3.6 Flash | 88.1 |
| GPT-5.6 Terra | 87.8 |
| GPT-5.4 | 87.5 |
| Gemini 3.5 Flash | 85.1 |
| GLM-5.2 | 79.2 |
| Qwen3.7-Plus | 74.3 |
| Gemini 3.5 Flash-Lite | 73.9 |
| Grok 4.3 | 69.6 |
| GLM-5.1 | 55.4 |
| Laguna M.1 | 54.1 |
| Kimi K2.6 | 51.8 |
| Laguna XS.2 | 31.9 |
| Inkling | 7.6 |