Legal Research Bench Administrative/Regulatory
Professional Work · 2026-06-23
This row reports the practice-area score for Administrative/Regulatory tasks in Vals AI Legal Research Bench.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Argon | 68.2 |
| Claude Opus 5 | 65.2 |
| Gemini 4 Carbon | 65.2 |
| Ember-1 | 63.6 |
| Kimi K3 | 63.6 |
| Muse Wonder Full | 63.6 |
| Muse Wonder | 63.6 |
| GLM-5.3 | 62.1 |
| Grok 4.6 | 62.1 |
| GPT-5.6 Sol | 60.6 |
| Muse Spark 1.3 | 60.6 |
| DeepSeek V4.1 Flash 1017 | 59.1 |
| MiMo-V2.6-Pro | 59.1 |
| Qwen3.8-Max | 59.1 |
| GLM-5.3 Flash | 57.6 |
| GPT-5.6 Terra | 57.6 |
| Kimi K3.1 Pro | 56.1 |
| Sonnet 5 | 54.5 |
| Grok 4.7 | 54.5 |
| Hy4 Preview | 54.5 |
| DeepSeek V4.1 Flash | 53.0 |
| GPT-5.5 | 53.0 |
| GPT-6.1 Sol | 53.0 |
| GPT-6 Astra | 53.0 |
| Grok 4.5 | 51.5 |
| Claude Haiku 5.5 | 50.0 |
| DeepSeek V4 Pro 0813 | 50.0 |
| Muse Spark Open | 50.0 |
| Qwen3.8 Max (0902) | 50.0 |
| GPT-5.6 Luna | 48.5 |
| Muse Spark 1.2 | 48.5 |
| Step 5 Preview | 48.5 |
| Gemini 3.8 Flash | 47.0 |
| MiMo-V2.6-Flash | 47.0 |
| Opus 4.7 | 45.5 |
| GPT-6.1 Terra | 45.5 |
| Qwen3.8 27B | 45.5 |
| Opus 4.8 | 43.9 |
| Sonnet 4.6 | 43.9 |
| Gemini 3.5 Flash | 43.9 |