Legal Research Bench Weighted Pass Business/Commercial
Professional Work · 2026-06-23
Weighted-pass practice-area score for Business/Commercial tasks in Vals AI Legal Research Bench.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5 | 92.2 |
| Claude Fable 5 | 91.4 |
| Muse Spark 1.1 | 89.1 |
| Kimi K3 | 89.1 |
| GPT-5.6 Sol | 89.0 |
| Sonnet 4.6 | 88.0 |
| GLM-5.2 | 87.5 |
| GPT-5.6 Luna | 87.3 |
| Grok 4.5 | 86.8 |
| GPT-5.5 | 86.8 |
| Sonnet 5 | 86.8 |
| DeepSeek-V4-Pro | 86.5 |
| Opus 4.8 | 86.1 |
| Gemini 3.6 Flash | 84.6 |
| GPT-5.6 Terra | 84.3 |
| Gemini 3.5 Flash | 84.1 |
| MiniMax M3 | 82.7 |
| GLM-5.1 | 81.7 |
| Qwen3.7-Max | 81.0 |
| Gemini 3.1 Pro Preview | 80.8 |
| Inkling | 79.1 |
| Kimi K2.6 | 78.4 |
| Inkling-Small | 78.2 |
| Grok 4.3 | 75.5 |
| Gemini 3.5 Flash-Lite | 73.2 |
| GPT-5.4 Mini | 71.2 |
| Laguna M.1 | 62.7 |
| Laguna XS.2 | 59.7 |