Legal Research Bench Weighted Pass Civil Litigation
Professional Work · 2026-06-23
Weighted-pass practice-area score for Civil Litigation tasks in Vals AI Legal Research Bench.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Carbon | 96.5 |
| Gemini 4 Argon | 95.1 |
| Muse Wonder Full | 92.9 |
| Claude Opus 5 | 92.8 |
| Gemini 3.8 Flash | 91.5 |
| Muse Spark 1.3 | 91.2 |
| Grok 4.6 | 91.1 |
| Hy4 Preview | 90.8 |
| Opus 4.8 | 90.8 |
| MiniMax M3.1 Flash | 90.7 |
| GPT-5.6 Sol | 90.2 |
| Grok 4.7 | 90.0 |
| Step 5 Preview | 90.0 |
| GLM-5.3 | 89.9 |
| GLM-5.3 Flash | 89.6 |
| Muse Wonder | 89.0 |
| Muse Spark 1.1 | 89.0 |
| Gemini 3.7 Flash | 88.9 |
| Grok 4.5 | 88.7 |
| Sonnet 5 | 88.7 |
| DeepSeek-V4-Flash-0731 | 88.7 |
| GPT-6 Astra | 88.7 |
| Muse Spark Open | 88.3 |
| Opus 4.7 | 88.2 |
| GPT-6.1 Sol | 87.7 |
| MiMo-V2.6-Pro | 87.7 |
| GPT-5.5 | 87.6 |
| GPT-5.6 Terra | 87.5 |
| Sonnet 4.6 | 87.4 |
| Muse Spark 1.2 | 87.4 |
| Ember-1 | 87.1 |
| DeepSeek V4.1 Flash | 87.1 |
| GLM-5.2 | 87.0 |
| Gemini 3.5 Flash | 86.3 |
| DeepSeek-V4-Pro | 86.0 |
| Mistral Large 4 Preview | 85.8 |
| DeepSeek V4 Pro 0813 | 85.8 |
| Gemini 3.6 Flash | 85.3 |
| Inkling | 85.0 |
| MiniMax M3 | 84.9 |