Qeravio
Canonical AI event

FSDP2 activation memory reduced for long sequences in v1.15.0

FSDP2 activation checkpointing now wraps the entire transformer layer, reducing saved activations.

9 Sept 20261 verified claims1 sources1 observations
What happened

The official source reports this update: v1.15.0: FSDP2 activation memory, dtensor improvements. v1.15.0: FSDP2 activation memory, dtensor improvements FSDP2 A large batch of FSDP2 work this release: two fixes that cut activation memory at long sequence lengths, tied-embedding support on torch >= 2.13, and a round of checkpointing correctness and scale fixes. Activation checkpointing was wrapping each child of the matched layer ( self_attn , mlp , the norms) instead of the layer itself, so every inter-child activation stayed saved for backward. It now wraps the layer.

Why it matters

This official update documents a development concerning v1.15.0: FSDP2 activation memory, dtensor improvements. Its practical significance depends on the scope and evidence stated by the source.

What to watch next

Read the official source update and verify its stated scope, evidence and timing before acting on it.

Connected knowledge

Entities affected by this event

Evidence trail

Sources behind the event