lynnreal_omni_flash_int8_lite: Requesting support for LoRA.
I tested this model and found it to be quite fast—almost twice the speed of the Base model—but it doesn't seem compatible with various LoRAs designed for H3;
The LoRA loader in ComfyUI throws an error and fails to work;
Could you please update it to support ComfyUI LoRAs? Thanks!
[WARNING] lora key not loaded: diffusion_model.blocks.42.attn.out_proj.alpha [WARNING] lora key not loaded: diffusion_model.blocks.42.attn.qkv_proj.alpha [WARNING] lora key not loaded: diffusion_model.blocks.42.mlp.fc1.alpha [WARNING] lora key not loaded: diffusion_model.blocks.42.mlp.fc2.alpha [WARNING] lora key not loaded: diffusion_model.blocks.43.attn.out_proj.alpha [WARNING] lora key not loaded: diffusion_model.blocks.43.attn.qkv_proj.alpha
lynnreal_omni_flash_int8_lite.safetensors