Legal Research Bench Immigration
Professional Work · 2026-06-23
This row reports the practice-area score for Immigration tasks in Vals AI Legal Research Bench.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Argon | 54.5 |
| Grok 4.6 | 54.5 |
| Muse Spark 1.2 | 54.5 |
| DeepSeek V4 Pro 0813 | 45.5 |
| Gemini 4 Carbon | 45.5 |
| GLM-5.3 Flash | 45.5 |
| GPT-5.6 Sol | 45.5 |
| GPT-6 Luna | 45.5 |
| Grok 4.7 | 45.5 |
| Kimi K3 | 45.5 |
| Muse Spark 1.3 | 45.5 |
| Muse Wonder | 45.5 |
| Qwen3.8-Max | 45.5 |
| Claude Haiku 5.5 | 36.4 |
| Opus 4.7 | 36.4 |
| Sonnet 5 | 36.4 |
| DeepSeek V4.1 Flash 1017 | 36.4 |
| DeepSeek V4.1 Flash | 36.4 |
| Gemini 3.7 Flash | 36.4 |
| GPT-5.5 | 36.4 |
| GPT-6 Astra | 36.4 |
| GPT-6 Sol | 36.4 |
| Grok 4.5 | 36.4 |
| Inkling | 36.4 |
| Muse Wonder Full | 36.4 |
| Step 5 Preview | 36.4 |
| Haiku 4.5 | 27.3 |
| Opus 4.8 | 27.3 |
| Claude Opus 5 | 27.3 |
| Sonnet 4.6 | 27.3 |
| DeepSeek-V4-Flash-0731 | 27.3 |
| Ember-1 | 27.3 |
| Gemini 3.8 Flash | 27.3 |
| Gemini 3 Flash Preview | 27.3 |
| GLM-5.2 | 27.3 |
| GLM-5.3 | 27.3 |
| Hy4 Preview | 27.3 |
| Kimi K2.5 | 27.3 |
| MiMo-V2.5-Pro | 27.3 |
| MiMo-V2.6-Pro | 27.3 |