See pre-training.
Pre-training The field’s word for the training run itself, in which a model learns from its data, such as human text. Post-training is the work done on the model after it, such as reinforcement learning. Chapter 2.
From The AI Compute Race. Read the book
Sources that use it
- SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD, arXiv:2607.20145, submitted 22 July 2026, revised 19 August 2026 (post-training, not pre-training from scratch: an end-to-end optimization practice on the Ascend NPU SuperPOD with the DeepSeek-V4 family as the workload — 34.22 percent model FLOPs utilization, “a 2.93x improvement over the open-source baseline recipe” — and a continued-pre-training and SFT workflow on that infrastructure). In the Sources
Related terms
From the glossary of The AI Compute Race.
How to cite this page
Dryburgh, Lee S. “Post-training.” In The AI Compute Race. DRYBURGH. https://dryburgh.com/vol1/glossary/post-training (accessed 2026-11-01).
BibTeX
@misc{dryburgh-post-training-2026,
author = {Dryburgh, Lee S.},
title = {{Post-training}},
howpublished = {In \emph{The AI Compute Race}, DRYBURGH},
year = {2026},
url = {https://dryburgh.com/vol1/glossary/post-training},
note = {Accessed 2026-11-01}
}