Legal Agent Benchmark held-out problem set; standard all-criteria-pass task success.
条件
1,235 problems tested; 16/1,251 removed for data defects before testing; average over 5 trials; internal harness uses bash and Python only with Sonnet 4.6 judge.
Legal Agent Benchmark held-out problem set; mean rubric-criterion pass rate.
条件
1,235 problems tested; average over 5 trials; internal harness uses bash and Python only with Sonnet 4.6 judge. Metric is a proposed extension to the held-out catalog entry.
Claude Opus 4.8 · Legal Agent Benchmark · Opus 5 官方发布表所列 held-out set · Scores and configurations
Claude Opus 4.8 record for Legal Agent Benchmark · Opus 5 官方发布表所列 held-out set. 2 score records for this release, including 2 exact values (may include historical records). Includes 2 source locations.
Legal Agent Benchmark held-out problem set; standard all-criteria-pass task success.
Conditions
1,235 problems tested; 16/1,251 removed for data defects before testing; average over 5 trials; internal harness uses bash and Python only with Sonnet 4.6 judge.
Legal Agent Benchmark held-out problem set; mean rubric-criterion pass rate.
Conditions
1,235 problems tested; average over 5 trials; internal harness uses bash and Python only with Sonnet 4.6 judge. Metric is a proposed extension to the held-out catalog entry.