Is there an existing issue for this?
What should this feature add?
Load and apply LoRAs for ERNIE-Image. Today there is no LoRA config for base ernie-image, and the LoRA loader raises
"Unsupported LoRA base model" (invokeai/backend/model_manager/load/model_loaders/lora.py), so community LoRAs
for this family cannot be installed or used.
Alternatives
None within InvokeAI.
Additional Content
Needs a LoRA config (LyCORIS/kohya and, where published, diffusers/PEFT key layouts), a key converter to the
model's module names, patching in the denoise node (sidecar patching for quantized and GGUF builds), and a LoRA
slot in Generate.
The full matrix of what loads today is in #9726 (Model Format Support page).
Is there an existing issue for this?
What should this feature add?
Load and apply LoRAs for ERNIE-Image. Today there is no LoRA config for base
ernie-image, and the LoRA loader raises"Unsupported LoRA base model" (
invokeai/backend/model_manager/load/model_loaders/lora.py), so community LoRAsfor this family cannot be installed or used.
Alternatives
None within InvokeAI.
Additional Content
Needs a LoRA config (LyCORIS/kohya and, where published, diffusers/PEFT key layouts), a key converter to the
model's module names, patching in the denoise node (sidecar patching for quantized and GGUF builds), and a LoRA
slot in Generate.
The full matrix of what loads today is in #9726 (Model Format Support page).