Lincoln Stein 532f82cb97

Optimize RAM to VRAM transfer (#6312) ...

* avoid copying model back from cuda to cpu

* handle models that don't have state dicts

* add assertions that models need a `device()` method

* do not rely on torch.nn.Module having the device() method

* apply all patches after model is on the execution device

* fix model patching in latents too

* log patched tokenizer

* closes #6375

---------

Co-authored-by: Lincoln Stein <lstein@gmail.com>

2024-05-24 17:06:09 +00:00

..

app

Optimize RAM to VRAM transfer (#6312)

2024-05-24 17:06:09 +00:00

assets

feat(api): chore: pydantic & fastapi upgrade

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

backend

Optimize RAM to VRAM transfer (#6312)

2024-05-24 17:06:09 +00:00

configs

feat(mm): support sdxl ckpt inpainting models

2024-04-28 12:57:27 +10:00

frontend

fix(ui): isLocal erroneously hardcoded

2024-05-25 00:05:44 +10:00

invocation_api

Remove support for Prompt-to-Prompt cross-attention control (aka .swap()). This feature is not widely used. It does not work with SDXL and is incompatible with IP-Adapter and regional prompting. The implementation is also intertwined with both text embedding and the UNet attention layers, resulting in a high maintenance burden. For all of these reasons, we have decided to drop support.

2024-04-09 10:57:02 -04:00

version

Update invokeai_version.py

2024-05-24 08:00:03 +10:00

__init__.py

Various fixes

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