Scaling Model Validation on BrowseComp-Plus Out-of-sample (val)
0.071MAEpredictive framework for agentic scaling
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| predictive framework for agentic scalingModel=GPT-5.2, Intelligence Index=752025.12 | 0.071 | 15.8 | 0.045 | — | 0.2 |