mirror of
https://github.com/invoke-ai/InvokeAI.git
synced 2026-01-14 23:08:05 -05:00
This corrects a regression in loading of these models due to a change of the embedding_manager parameter `num_vectors_per_token` Fixes #1718