troyhan's picture
Update results.csv
09686b9 verified
Raw History Blame Contribute Delete
641 Bytes
Model ID,Compilation Success,Partial Replication,Correct Coefficient Direction,Significant Level Correctness
GPT 5.5 - Python,36.7,12.1,30.7,22.2
GPT 5.5 - Stata,35.9,6.5,21.0,11
GPT 5.5 - MetaGPT,55.0,11.4,48.5,32.5
GPT 5.5 - MetricsAI,52.1,14.2,46.5,33.6
GPT 5.5 - Codex,99.5,26.4,86.5,62
DeepSeek v4 Pro - MetaGPT,67.4,14.4,57.8,38.2
Deepseek v4 Pro - MetricsAI,13.2,5.7,13.2,10.1
Deepseek v4 Pro - Python,25.3,8.4,21,15.9
Deepseek v4 Pro - Stata,20.3,8.1,17.2,14
Deepseek v4 Pro - Codex,86.3,32.8,69.3,53.6
GPT 4o - MetaGPT,67.4,22,57.8,36.8
GPT 4o - MetricsAI,84.5,22.7,61,40.5
GPT 4o - Python,7,3.4,5.9,4.9
GPT 4o - Stata,1.1,0.2,1,0.6