TestsBenchmarks · Knowledge
PRBench Legal (Scale)
Expert-rubric graded professional legal reasoning questions (Scale SEAL).
% solvedHigher is betterCounts toward:Weights: Research & analysis ×0.7The test's websiteOfficial page ↗
Measured by independent testersIndependent Reported by the makerVendor-reported
| Thinking levelSetting | |||||||
|---|---|---|---|---|---|---|---|
| 1 | Muse Spark 1.3 | Meta (Meta Superintelligence Labs, Muse) | 61.6% | DefaultMuse Spark 1.3 | Independent testIndependentScale AI (SEAL) ↗ | — | 14 Sep 2026 |
| 2 | Muse Spark 1.1 | Meta (Meta Superintelligence Labs, Muse) | 57.0% | DefaultMuse Spark 1.1 | Independent testIndependentScale AI (SEAL) ↗ | — | 9 Jul 2026 |
| 3 | Claude Fable 5 | Anthropic | 52.6% | Defaultclaude-fable 5 | Independent testIndependentScale AI (SEAL) ↗ | — | 28 Jul 2026 |
| 4 | Muse Spark | Meta (Meta Superintelligence Labs, Muse) | 52.3% | DefaultMuse Spark | Independent testIndependentScale AI (SEAL) ↗ | — | 8 Apr 2026 |
| 5 | Claude Opus 4.6 | Anthropic | 52.3% | No reasoningclaude-opus-4-6 (Non-Thinking) | Independent testIndependentScale AI (SEAL) ↗ | — | 17 Feb 2026 |
| 6 | Claude Fable 5.1 | Anthropic | 51.6% | DefaultFable 5.1 | Independent testIndependentScale AI (SEAL) ↗ | — | 3 Sep 2026 |
| 7 | GPT-5.6 Sol | OpenAI | 50.5% | Maxgpt-5.6-sol (max) | Independent testIndependentScale AI (SEAL) ↗ | — | 28 Jul 2026 |
| 8 | GPT-5.1 | OpenAI | 49.3% | Defaultgpt-5.1-thinking | Independent testIndependentScale AI (SEAL) ↗ | — | 11 Dec 2025 |
| 9 | Gemini 3.8 Flash | Google (Gemini / DeepMind) | 48.9% | DefaultGemini 3.8 Flash | Independent testIndependentScale AI (SEAL) ↗ | — | 9 Sep 2026 |
| 10 | GPT-6 Astra | OpenAI | 48.4% | DefaultGPT 6 Astra | Independent testIndependentScale AI (SEAL) ↗ | — | 9 Sep 2026 |
| 11 | Grok 4.7 | SpaceXAI (formerly xAI) | 47.6% | Extra highGrok 4.7 (xHigh) | Independent testIndependentScale AI (SEAL) ↗ | — | 25 Sep 2026 |
| 12 | Claude Opus 5.5 | Anthropic | 47.6% | Defaultclaude-opus-5-5 | Independent testIndependentScale AI (SEAL) ↗ | — | 25 Sep 2026 |
| 13 | GPT-5.2 Pro | OpenAI | 45.4% | Defaultgpt-5.2-pro-2025-12-11 | Independent testIndependentScale AI (SEAL) ↗ | — | 16 Dec 2025 |
| 14 | GPT-5.4 | OpenAI | 44.4% | Highgpt-5.4 (High) | Independent testIndependentScale AI (SEAL) ↗ | — | 22 Apr 2026 |
| 15 | Claude Opus 4.5 | Anthropic | 44.2% | Defaultclaude-opus-4-5-20251101-thinking | Independent testIndependentScale AI (SEAL) ↗ | — | 26 Nov 2025 |
| 16 | Gemini 3.1 Pro (Preview) | Google (Gemini / DeepMind) | 44.0% | Defaultgemini-3.1-pro | Independent testIndependentScale AI (SEAL) ↗ | — | 23 Mar 2026 |
| 17 | Kimi K2.5 | Moonshot AI (Kimi) | 43.8% | Defaultkimi-k2.5 | Independent testIndependentScale AI (SEAL) ↗ | — | 13 Feb 2026 |
| 18 | GPT-6 Luna | OpenAI | 42.9% | Defaultgpt-6-luna | Independent testIndependentScale AI (SEAL) ↗ | — | 25 Sep 2026 |
| 19 | GPT-6 Sol | OpenAI | 38.6% | Defaultgpt-6-sol | Independent testIndependentScale AI (SEAL) ↗ | — | 25 Sep 2026 |