Legal Research Bench Weighted Pass Criminal
Professional Work · 2026-06-23
Weighted-pass practice-area score for Criminal tasks in Vals AI Legal Research Bench.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Carbon | 90.6 |
| Gemini 4 Argon | 90.1 |
| Muse Wonder Full | 89.4 |
| Gemini 3.8 Flash | 88.8 |
| Claude Opus 5 | 88.1 |
| Grok 4.7 | 87.9 |
| Opus 4.8 | 87.7 |
| Muse Wonder | 87.4 |
| GLM-5.3 | 87.2 |
| Grok 4.6 | 87.2 |
| MiMo-V2.6-Pro | 87.2 |
| Muse Spark 1.3 | 86.7 |
| GPT-5.6 Sol | 86.4 |
| GLM-5.3 Flash | 85.9 |
| Sonnet 4.6 | 85.8 |
| Muse Spark Open | 85.5 |
| Muse Spark 1.1 | 85.4 |
| Qwen3.8-Max | 84.9 |
| Kimi K3 | 84.6 |
| GPT-6.1 Sol | 84.5 |
| Gemini 3.7 Flash | 84.3 |
| Step 5 Preview | 84.1 |
| Sonnet 5 | 84.1 |
| Claude Haiku 5.5 | 83.8 |
| DeepSeek V4 Pro 0813 | 83.6 |
| Mistral Large 4 Preview | 83.5 |
| DeepSeek V4.1 Flash 1017 | 83.2 |
| GPT-6 Astra | 83.2 |
| Muse Spark 1.2 | 83.1 |
| Grok 4.5 | 82.6 |
| Ember-1 | 82.6 |
| GPT-5.6 Terra | 82.6 |
| Gemini 3.5 Flash | 82.4 |
| GPT-5.6 Luna | 82.2 |
| Opus 4.7 | 82.0 |
| Qwen3.8 Max (0902) | 81.9 |
| GPT-5.5 | 81.8 |
| Hy4 Preview | 81.8 |
| Mercury 3 | 81.6 |
| Gemini 3.6 Flash | 81.0 |