Qeravio
Canonical AI event

TRL v1.14.0 drops six experimental trainers, adds Triton kernel for log-probs and entropy.

The v1.14.0 release of TRL removes six experimental trainers, including GRPOWithReplayBufferTrainer and BCOTrainer.

25 Sept 20261 verified claims1 sources1 observations
What happened

The official source reports this update: v1.14.0. Features trl.losses is gone: DPO, KTO and GRPO stream their own log-probs Warning from trl.losses import FusedLinearDPOLoss (or FusedLinearKTOLoss , FusedLinearGRPOLoss , FusedLinearJSDLoss ) no longer works. The module introduced in v1.13 has been removed. v1.13 vendored Liger's chunked_loss into trl.losses as a holding action, not a destination ( #7063 ). The problem it was holding: those classes reimplement each trainer's loss math, so TRL carried two copies of every formula, and copies drift. That drift is where the bugs were.

Why it matters

This official update documents a development concerning v1.14.0. 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