Lincoln Stein
Lincoln Stein
@brandonrising Are you still working on this?
I'm just wondering if this PR is still actively being worked on?
Wheel of commit [589a795](https://github.com/invoke-ai/InvokeAI/pull/5997/commits/589a7959c019bc56e23ad3d989d015e443ffa20b) [InvokeAI-4.2.3-py3-none-any.whl.zip](https://github.com/user-attachments/files/15527851/InvokeAI-4.2.3-py3-none-any.whl.zip)
@raldone01 Thank you so much for giving the PR a try and your valuable feedback. I think I know where the meta tensor bug is occurring and should have a...
@raldone01 I've fixed what I believe to be the bug with changing models. Unfortunately I don't have access to a multi-GPU system at the moment, and have only tested it...
> I tested your branch again. Model switching still errors out but only sometimes, with `Cannot copy out of meta tensor; no data!` but invoke can recover without having to...
@raldone01 I just committed a series of changes that should make the multi-GPU support more stable. If you have a chance to check it out, let me know how it...
**This PR is no longer being maintained. For InvokeAI Multi-GPU support, please see https://github.com/lstein/InvokeAI-MGPU**
> I ran into a bug during testing. I hadn't thought about this before, but this approach breaks if a model is moved between devices while a patch is applied....
@RyanJDick I finally got back to this after an interlude. It was a relatively minor fix to get all the model patching done after loading the model into the target...