Legal Benchmarks - Contract Workflows - Task Pass Rate: leaderboard
Legal Benchmarks (legalbenchmarks.ai) model leaderboard, Contract Workflows: lawyer-authored drafting and review tasks completed in a sealed agentic harness, two attempts each, high reasoning effort where offered. Share of tasks meeting every applicable criterion in both attempts, judged per criterion by GLM 5.3 Flash; %, higher is better.
Metric: Task pass rate (%, every applicable criterion met in both attempts). Source: www.legalbenchmarks.ai. Saturation forecast: Around 2029. 26 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 48.4 | #3 |
| 2 | Claude Fable 5.1 | 48.4 | #10 |
| 3 | Muse Spark 1.3 | 48.4 | #31 |
| 4 | Claude Opus 5 | 45.2 | #16 |
| 5 | Qwen 3.8 Flash | 38.7 | #62 |
| 6 | Claude Haiku 5.5 | 35.5 | #43 |
| 7 | Qwen 3.8 Max | 29 | #34 |
| 8 | Claude Sonnet 5.5 | 29 | #14 |
| 9 | Gemini 3.8 Flash | 25.8 | #30 |
| 10 | GPT-6 Astra | 25.8 | #6 |
| 11 | GPT-6.1 Sol | 25.8 | #7 |
| 12 | Gemini 4 Argon | 25.8 | #9 |
| 13 | Mistral Large 4 | 25.8 | #106 |
| 14 | Grok 4.6 | 25.8 | #29 |
| 15 | GPT-5.6 Sol | 22.6 | #22 |
Interactive version: theaggregate.ai/benchmark?slug=legal-benchmarks-contract-workflows-task-pass-rate · How It Works · Data refreshed daily, snapshot 2026-11-05.