Legal Research Bench Criminal
Professional Work · 2026-06-23
This row reports the practice-area score for Criminal tasks in Vals AI Legal Research Bench.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Carbon | 59.6 |
| Gemini 4 Argon | 56.1 |
| Muse Wonder Full | 49.1 |
| Grok 4.7 | 47.4 |
| Muse Spark 1.3 | 47.4 |
| Opus 4.8 | 45.6 |
| Claude Opus 5 | 43.9 |
| GPT-5.6 Sol | 43.9 |
| MiMo-V2.6-Pro | 43.9 |
| Muse Wonder | 43.9 |
| Qwen3.8 Max (0902) | 43.9 |
| Claude Haiku 5.5 | 42.1 |
| Grok 4.6 | 42.1 |
| Sonnet 5 | 40.4 |
| Gemini 3.8 Flash | 40.4 |
| Qwen3.8-Max | 40.4 |
| GLM-5.3 | 38.6 |
| Muse Spark 1.2 | 38.6 |
| Step 5 Preview | 38.6 |
| Sonnet 4.6 | 36.8 |
| GPT-5.5 | 36.8 |
| Grok 4.5 | 36.8 |
| Hy4 Preview | 36.8 |
| Kimi K3 | 36.8 |
| Muse Spark 1.1 | 36.8 |
| DeepSeek V4.1 Flash 1017 | 35.1 |
| DeepSeek V4 Pro 0813 | 35.1 |
| GPT-5.6 Terra | 35.1 |
| GPT-6.1 Sol | 35.1 |
| Muse Spark Open | 35.1 |
| DeepSeek V4.1 Flash | 33.3 |
| Gemini 3.7 Flash | 33.3 |
| GLM-5.3 Flash | 33.3 |
| Ember-1 | 31.6 |
| Mistral Large 4 Preview | 31.6 |
| Opus 4.7 | 29.8 |
| GLM-5.1 | 29.8 |
| GPT-5.6 Luna | 29.8 |
| GPT-6 Astra | 29.8 |
| Mercury 3 | 29.8 |