Harvey LAB-AA
A benchmark for evaluating the performance of AI on tasks in the legal domain.
Timeline (last 90 days)
In Meta's own Agent benchmark, Muse Wonder scored 10.0 on the Harvey LAB-AA legal task, and Claude Opus 5.5 scored 4.2.
In Harvey LAB-AA legal analysis, Muse Wonder achieved 13.1%, more than three times Claude Opus 5.5's 4.2%; in Vals Finance Agent V2 it ranked first with 59.1%.
On DeepSWE 1.1, Muse Wonder beats Claude Opus 5.5 78.8 to 74.2, leads 10.0 to 4.2 on the Harvey legal agent benchmark LAB-AA, and leads 53.1 to 46.3 on SWE-Marathon.
All Facts (3)
In Meta's own Agent benchmark, Muse Wonder scored 10.0 on the Harvey LAB-AA legal task, and Claude Opus 5.5 scored 4.2.
50%UnverifiedIn Harvey LAB-AA legal analysis, Muse Wonder achieved 13.1%, more than three times Claude Opus 5.5's 4.2%; in Vals Finance Agent V2 it ranked first with 59.1%.
50%UnverifiedOn DeepSWE 1.1, Muse Wonder beats Claude Opus 5.5 78.8 to 74.2, leads 10.0 to 4.2 on the Harvey legal agent benchmark LAB-AA, and leads 53.1 to 46.3 on SWE-Marathon.
50%