Legal Research Bench Weighted Pass Family
Professional Work · 2026-06-23
Weighted-pass practice-area score for Family tasks in Vals AI Legal Research Bench.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5 | 86.6 |
| Kimi K3 | 79.3 |
| GPT-5.6 Luna | 78.4 |
| Claude Fable 5 | 77.9 |
| Opus 4.8 | 77.3 |
| Sonnet 4.6 | 76.8 |
| GPT-5.6 Sol | 76.7 |
| Gemini 3.5 Flash | 74.9 |
| GPT-5.6 Terra | 74.2 |
| Muse Spark 1.1 | 73.7 |
| Sonnet 5 | 72.2 |
| Inkling | 71.7 |
| GPT-5.5 | 70.8 |
| Grok 4.5 | 70.6 |
| GLM-5.2 | 69.3 |
| Inkling-Small | 68.9 |
| MiniMax M3 | 68.0 |
| DeepSeek-V4-Pro | 66.3 |
| Kimi K2.6 | 61.0 |
| Gemini 3.6 Flash | 59.6 |
| Qwen3.7-Max | 59.3 |
| Gemini 3.1 Pro Preview | 58.2 |
| GLM-5.1 | 58.2 |
| GPT-5.4 Mini | 51.6 |
| Gemini 3.5 Flash-Lite | 49.7 |
| Grok 4.3 | 47.3 |
| Laguna M.1 | 42.0 |
| Laguna XS.2 | 30.2 |