Ryan Dick c738fe051f Split T5 encoding and CLIP encoding into separate functions to ensure that all model references are locally-scoped so that the two models don't have to be help in memory at the same time.

2024-08-29 19:08:18 +00:00

..

app

Split T5 encoding and CLIP encoding into separate functions to ensure that all model references are locally-scoped so that the two models don't have to be help in memory at the same time.

2024-08-29 19:08:18 +00:00

assets

feat(api): chore: pydantic & fastapi upgrade

2023-10-17 14:59:25 +11:00

backend

Fix bug in InvokeLinear8bitLt that was causing old state information to persist after loading from a state dict. This manifested as state tensors being left on the GPU even when a model had been offloaded to the CPU cache.

2024-08-29 19:08:18 +00:00

configs

Remove dependency on flux config files

2024-08-26 20:17:50 -04:00

frontend

feat(ui): add fields for CLIP embed models and Flux VAE models in workflows

2024-08-29 11:52:51 -04:00

invocation_api

Added support for bounding boxes in the Invocation API

2024-08-26 08:03:30 +10:00

version

chore: bump version to v4.2.9rc1

2024-08-27 12:09:27 -04:00

__init__.py

Various fixes

2023-01-30 18:42:17 -05:00