Average token usage per workflow node, based on the latest completed run for each paper.
| Node | Papers | Avg Input | Avg Output | Avg Total |
|---|---|---|---|---|
| Code_Availability_Check | 318 | 1,671 (±1,861) | 99 (±42) | 1,771 (±1,874) |
| Code_Embedding | 74 | 25,770 (±21,953) | 110 (±108) | 25,881 (±22,005) |
| Code_Repository_Analysis | 74 | 75,186 (±17,588) | 1,406 (±329) | 76,593 (±17,643) |
| Dataset_Documentation_Check | 50 | 13,204 (±1,548) | 1,018 (±159) | 14,222 (±1,567) |
| Final_Aggregation | 115 | 919 (±92) | 482 (±46) | 1,401 (±100) |
| Paper_Type_Classification | 115 | 5,009 (±547) | 238 (±36) | 5,248 (±553) |
| Reproducibility_Checklist | 115 | 27,579 (±4,181) | 2,146 (±328) | 29,725 (±4,185) |
| Section_Embeddings | 115 | 4,700 (±451) | 0 | 4,700 (±451) |