LLM Stats (Legal Agent Benchmark): leaderboard
LLM Stats mirror of the Legal Agent Benchmark, evaluating legal-agent task completion over closed-universe matter files with expert rubrics.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around 2029. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 13.3 | #16 |
| 2 | Claude Opus 5 | 11.7 | #12 |
| 3 | Claude Sonnet 5.5 | 11.7 | #11 |
| 4 | Claude Opus 5.5 | 8.3 | #3 |
| 5 | Claude Opus 4.7 | 7.1 | #49 |
| 6 | Claude Sonnet 5 | 5.8 | #57 |
| 7 | Claude Sonnet 4.6 | 5.4 | #90 |
| 8 | Claude Opus 4.6 | 4.2 | #63 |
| 9 | GPT-5.5 | 2.1 | #31 |
| 10 | Gemini 3.5 Flash | 0.8 | #60 |
| 11 | GPT-5.4 | 0.4 | #79 |
| 12 | Gemini 3.1 Pro (Preview) | 0 | #58 |
| 13 | Gemini 3 Flash | 0 | #99 |
| 14 | GPT-5.4 Mini | 0 | #212 |
| 15 | Gemini 3.1 Flash Lite | 0 | #191 |
No result here: #6 GPT-6 Astra, #7 GPT-6.1 Sol, #8 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-legal-agent-benchmark · How It Works · Data refreshed daily, snapshot 2026-10-20.