Legal AI Intelligence
Rankings by a 0–100 composite across 6 weighted capabilities for legal work. Source: Artificial Analysis Legal Index.
| # | Model | Maker | Legal Index | Strengths |
|---|---|---|---|---|
| 1 | Claude Fable 5 (with fallback) | Anthropic | 59 | |
| 2 | GPT-5.6 Sol (max) | OpenAI | 52 | |
| 3 | Grok 4.5 (high) | xAI | 51 | |
| 4 | Kimi K3 | Moonshot AI | 50 | |
| 5 | Claude Opus 4.8 (max) | Anthropic | 50 |
Scores are Artificial Analysis's published composite across 6 weighted capabilities; we do not recompute them. Maximum possible score: 100. Source: Artificial Analysis Legal Index as of July 20, 2026.
Artificial Analysis combines 6 weighted capabilities into one published Legal Index score:
General benchmark rankings (coding, math) don't predict legal performance. A model that tops HumanEval may underperform on contract analysis or bar exam tasks.
Drafting intake emails, analyzing contracts, and answering legal research questions each favor different model strengths. The leaderboard helps identify the right tool for each task.
New models release monthly. A model that was best-in-class 6 months ago may have been surpassed. Check the Artificial Analysis Legal Index regularly or talk to us about AI implementation for your firm.