#Can't convert checkpoints to diffusers

9 messages · Page 1 of 1 (latest)

dry deltaBOT
#

Hi @glossy gust. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:

#

I'm sorry to hear that you're having trouble with converting checkpoints to diffusers in Invoke AI. Based on the knowledge sources provided, there have been several issues reported on the Invoke AI GitHub page related to checkpoint conversion problems.

For instance, in issue #4711 (https://github.com/invoke-ai/InvokeAI/issues/4711), a user reported a validation error for CheckpointConfig config. In issue #4539 (https://github.com/invoke-ai/InvokeAI/issues/4539), a user reported an error when trying to generate images with SDXL models. In issue #4277 (https://github.com/invoke-ai/InvokeAI/issues/4277), a user reported that vpred SD 1.5 models don't convert to diffusers properly. In issue #5494 (https://github.com/invoke-ai/InvokeAI/issues/5494), a user reported that they could no longer generate images with any new SDXL models that they install. In issue #3561 (https://github.com/invoke-ai/InvokeAI/issues/3561), a user reported that they could not load a legacy checkpoint, and the diffusers conversion failed.

Unfortunately, without more specific error messages or logs, it's hard to diagnose the exact issue you're facing. However, you might find it helpful to review these issues and their comments to see if any of the solutions or workarounds suggested there might apply to your situation.

#

If you're still having trouble, I would recommend creating a new issue on the Invoke AI GitHub page (https://github.com/invoke-ai/InvokeAI/issues) with as much detail as possible about your problem, including any error messages or logs you have, your system configuration, and the specific steps you're taking when you encounter the problem. This will help the developers and other users to better understand and address your issue.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:

#

@warm plank kapa.ai is still learning and improving, please let us know how it did by reacting below

cold sail
#

@glossy gust how much RAM do you have?

glossy gust
#

11vram, 16ram

#

I just don't get it, cause I managed to create images few weeks ago. And one day, I had this convert issue

#

Updating to 4.04/install from sctach didn't fix it

#

So maybe I did something wrong, maybe I didn't reset my installation completly, I don't know what to do, how to check..