lllyasviel/ControlNet

Training still OOM on 8GB gpu.

Open

#21 geöffnet am 12. Feb. 2023

Auf GitHub ansehen
 (2 Kommentare) (3 Reaktionen) (0 zugewiesene Personen)Python (2.500 Forks)batch import
help wanted

Repository-Metriken

Stars
 (27.526 Stars)
PR-Merge-Metriken
 (Keine gemergten PRs in 30 T)

Beschreibung

It seems that I already used many tricks but training still OOM for 8GB gpu. But inference is good now.

This is strange because I know some textural inversion or dreambooth can be trained on 8GB.

What is the secrect of Automatic1111's optimization? Although xformers may help a bit, the currect sliced attention should require even smaller mem than xformers.

Does it make sence to move text encoder and vae outside gpu when training?

Contributor Guide