invoke-ai
[bug]: Model installation failed (fresh install of 4.2.0)
Is there an existing issue for this problem?
- [X] I have searched the existing issues
Operating system
Windows
GPU vendor
Nvidia (CUDA)
GPU model
No response
GPU VRAM
No response
Version number
4.2.0
Browser
Edge latest
Python dependencies
No response
What happened
Installation of the IP Adapter SDXL from the model manager fails.
What you expected to happen
Installation successful
How to reproduce the problem
- Fresh install InvokeAI 4.2.0 (latest)
- In the model manager, click the "+" button of IP Adapater SDXL in the starter models tab
- The installation fails (screenshot below)
The target folder is empty before installation.
When the downloading of the image_encoder finishes, InvokeAI auto creates the folder "any/clip_vision/ip_adapter_sdxl_image_encoder' with config.json and model.safetensors inside, then it reports this error:
**[ModelInstallService]::ERROR --> Model install error: InvokeAI/ip_adapter_sdxl_image_encoder
DuplicateModelException: A model with path 'any/clip_vision/ip_adapter_sdxl_image_encoder' is already installed**
Additional context
Attempting to use it would trigger another error:
invokeai.backend.model_manager.config.InvalidModelConfigException: Files for model 'ip_adapter_sd_image_encoder' not found
Discord username
No response
The installation error message is saying the SDXL IP Adapter image encoder is already installed - that's not a problem.
Then, it appears you are attempting to use a SD 1.5 IP adapter, but don't have the SD 1.5 encoder installed. Which IP adapter model are you attempting to use when you get that error message?
