Legal Agent Benchmark
Can the model produce useful work in a professional knowledge-work setting?
Knowledge Work4 ranked models4 reported values1 reportsHigher is better
Top rankings
One row per model, using its best reported score across effort settings.
| Rank | Model | Best score | Best reported setting |
|---|---|---|---|
| 1 | Claude Fable 5Anthropic | 13.3% | Reported configurationAnthropic report1 value · 1 reportJul 24, 2026 · source |
| 2 | Claude Opus 5Anthropic | 11.7% | max effortAnthropic report1 value · 1 reportJul 24, 2026 · source |
| 3 | Claude Opus 4.8Anthropic | 10.4% | Reported configurationAnthropic report1 value · 1 reportJul 24, 2026 · source |
| 4 | GPT-5.6 SolOpenAI | 2.5% | Reported configurationAnthropic report1 value · 1 reportJul 24, 2026 · source |
Effort curve
Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology
Can the model produce useful work in a professional knowledge-work setting?
Professional analysis, document, finance, office, or domain-specific tasks. Higher is better. The value is the percentage reported in this lab's table.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
Method / source