CRAD
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
CRAD
100Success Rate
22
CRAD Debt
100Success Rate
14
CRAD (test)
100Success Rate
7
CRAD Qwen2.5-3B-Instruct counterparty
85Success Rate
4
CRAD ChatGPT-4o-mini counterparty
95Success Rate
4
CRAD DeepSeek-V3 counterparty
100Success Rate
4