Full recomputation costs a third more compute to cut activation memory by an order of magnitude. The per-layer arithmetic that says whether you need it at all, the selective variant that costs almost nothing, and the point at which a smaller micro-batch is the cheaper way out.
Updated Sep 2026 · Grounded in real AI infrastructure interview loops and written to a senior-engineer editorial bar, with every number worked and every diagram hand-built.
Full recomputation costs a third more compute to cut activation memory by an order of magnitude. The per-layer arithmetic that says whether you need it at all, the selective variant that costs almost nothing, and the point at which a smaller micro-batch is the cheaper way out.
Ranked by how closely each one overlaps this question's topic, so the first card is the thing to read if the answer above moved too fast.
Scored on doing the activation memory per layer before naming a technique, on knowing that the recompute cost is a third and why, and on reaching for selective checkpointing before full.
No comments yet — be the first to share your approach.
