GMP Bench
← Back to Leaderboard

Claude Sonnet 5

AnthropicClosed
Overall Score
89.1%

65 evaluations

GMP Knowledge
100.0%

44 evals

Task Completion
86.4%

18 evals

Multimodal
76.8%

3 evals, scored separately

Avg Latency
31.1s

across all evals

Score by Category