Legal Benchmarks - Contract Workflows - Task Pass Rate: leaderboard
Metric: Task pass rate (%, every applicable criterion met in both attempts). Source: www.legalbenchmarks.ai. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5.1 | 48.4 |
| 2 | Muse Spark 1.3 | 48.4 |
| 3 | Claude Opus 5 | 45.2 |
| 4 | Qwen 3.8 Flash | 38.7 |
| 5 | Qwen 3.8 Max | 29 |
| 6 | GPT-6 | 25.8 |
| 7 | Grok 4.6 | 25.8 |
| 8 | Gemini 3.8 Flash | 25.8 |
| 9 | GPT-5.6 Sol | 22.6 |
| 10 | DeepSeek V4 Flash | 22.6 |
| 11 | Kimi K3 | 22.6 |
| 12 | GLM-5.3 Flash | 22.6 |
| 13 | DeepSeek V4 Pro | 19.4 |
| 14 | GPT-5.6 Luna | 19.4 |
| 15 | GLM-5.3 | 19.4 |
Interactive version: theaggregate.ai/benchmark?slug=legal-benchmarks-contract-workflows-task-pass-rate · How It Works · Data refreshed daily, snapshot 2026-09-19.