Legal Research Bench Weighted Pass Health
Professional Work · 2026-06-23
Weighted-pass practice-area score for Health tasks in Vals AI Legal Research Bench.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-5.6 Sol | 97.5 |
| Claude Opus 5 | 96.8 |
| Claude Fable 5 | 94.5 |
| GPT-5.6 Terra | 94.0 |
| GPT-5.6 Luna | 93.8 |
| GPT-5.5 | 93.6 |
| Grok 4.5 | 93.2 |
| Kimi K3 | 93.0 |
| MiniMax M3 | 92.2 |
| Sonnet 5 | 92.1 |
| Gemini 3.5 Flash | 91.5 |
| GLM-5.2 | 91.3 |
| DeepSeek-V4-Pro | 91.2 |
| Opus 4.8 | 90.8 |
| Sonnet 4.6 | 90.3 |
| Inkling | 88.5 |
| Muse Spark 1.1 | 87.9 |
| Gemini 3.6 Flash | 87.9 |
| Qwen3.7-Max | 86.7 |
| Inkling-Small | 84.4 |
| Gemini 3.1 Pro Preview | 83.4 |
| GLM-5.1 | 82.8 |
| Kimi K2.6 | 81.7 |
| Grok 4.3 | 81.2 |
| Gemini 3.5 Flash-Lite | 75.7 |
| GPT-5.4 Mini | 73.8 |
| Laguna M.1 | 69.2 |
| Laguna XS.2 | 62.8 |