AnalysisAI ModelsSeptember 17, 2026

Linum releases JiT-DDT, training text-to-image 3.6× faster

JiT-DDT trains a text-to-image model with 3.6× fewer GPU-hours than the Linum v2 baseline while generating images with 4× the pixels. Code and weights are Apache 2.0 licensed; Linum calls it a research artifact, not a full model release.

1 source

More stories today

Open the live feed