mirror of
https://github.com/invoke-ai/InvokeAI.git
synced 2026-02-07 03:45:24 -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