For verbal and visual learning, my advisor typically has me use either the Child and Adolescent Memory Profile (ChAMP; Sherman & Brooks, 2015), or the Wide Range Assessment of Memory and Learning (WRAML-2; Sheslow & Adams, 2003). Both involve immediate recall, delayed recall, and recognition subtests. However, both of these would be in-session delays of less than an hour.

Walhovd et al. (2020) published a paper in which they used alternate forms of the California Verbal Learning Test (CVLT-II; Delis et al., 2000) to quantify "learning slope", and then had participants engage in memory training trials with a written recall word list of 100 nouns. Participants were re-evaluated using these measures over 11 years. While I know their 100 noun list is not a rigorously validated measure like some of the previously mentioned test batteries, maybe their methods section might spur some insightful solution for you?

References

Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (2000). California verbal learning test: Second edition (CVLT-II). The Psychological Corporation.

Sherman, E. M. S., & Brooks, B. L. (2015). Child and adolescent memory profile: Professional manual. PAR, Inc.

Sheslow, D., & Adams, W. (2003). Wide range assessment of memory and learning (2nd ed.). PAR, Inc.

Walhovd, K. B., Bråthen, A. C. S., Panizzon, M. S., Mowinckel, A. M., Sørensen, Ø., de Lange, A.-M. G., Krogsrud, S. K., Håberg, A., Franz, C. E., Kremen, W. S., & Fjell, A. M. (2020). Within-session verbal learning slope is predictive of lifespan delayed recall, hippocampal volume, and memory training benefit, and is heritable. Scientific Reports, 10(21158). https://doi.org/10.1038/s41598-020-78225-1

I would agree with the two responses and add the DKEFS and NEPSY-II as well.