ValueError: Invalid LoRA checkpoint.

[AlgorithmicKing]

i am getting this error when trying to load lora model:

LoRA adapter name : proteus_v04beta
Traceback (most recent call last):
  File "I:\fastsdcpu-1.0.0-beta.30\env\lib\site-packages\gradio\queueing.py", line 501, in call_prediction
    output = await route_utils.call_process_api(
  File "I:\fastsdcpu-1.0.0-beta.30\env\lib\site-packages\gradio\route_utils.py", line 253, in call_process_api
    output = await app.get_blocks().process_api(
  File "I:\fastsdcpu-1.0.0-beta.30\env\lib\site-packages\gradio\blocks.py", line 1695, in process_api
    result = await self.call_function(
  File "I:\fastsdcpu-1.0.0-beta.30\env\lib\site-packages\gradio\blocks.py", line 1235, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "I:\fastsdcpu-1.0.0-beta.30\env\lib\site-packages\anyio\to_thread.py", line 56, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
  File "I:\fastsdcpu-1.0.0-beta.30\env\lib\site-packages\anyio\_backends\_asyncio.py", line 2144, in run_sync_in_worker_thread
    return await future
  File "I:\fastsdcpu-1.0.0-beta.30\env\lib\site-packages\anyio\_backends\_asyncio.py", line 851, in run
    result = context.run(func, *args)
  File "I:\fastsdcpu-1.0.0-beta.30\env\lib\site-packages\gradio\utils.py", line 692, in wrapper
    response = f(*args, **kwargs)
  File "I:\fastsdcpu-1.0.0-beta.30\src\frontend\webui\lora_models_ui.py", line 67, in on_click_load_lora
    load_lora_weight(
  File "I:\fastsdcpu-1.0.0-beta.30\src\backend\lora.py", line 65, in load_lora_weight
    pipeline.load_lora_weights(
  File "I:\fastsdcpu-1.0.0-beta.30\env\lib\site-packages\diffusers\loaders\lora.py", line 1248, in load_lora_weights
    raise ValueError("Invalid LoRA checkpoint.")
ValueError: Invalid LoRA checkpoint.

[rupeshs]

please check our tutorial https://youtu.be/N_LUxd6G2nk?si=1tCYvnkhvByelh54

[AlgorithmicKing]

same error

[rupeshs]

You need to use valid lora models

[AlgorithmicKing]

for example?

[AlgorithmicKing]

this is the model i used isnt it valid? https://civitai.com/models/267242/proteus

[AlgorithmicKing]

oh nvm i figured it out it was a checkpoint and not a lora model. so now can i use checkpoints in fastsd cpu?

[rupeshs]

yes it is not a lora model

[AlgorithmicKing]

can i use checkpoint models on fastsd cpu

[AlgorithmicKing]

is it true?: gemini: FastSD CPU does not currently support loading custom checkpoint models. It is designed to work with specific models optimized for faster inference on CPUs.

However, FastSD CPU does support:

LCM (Latent Consistency Models): These models are specifically designed for fast inference on CPUs. OpenVINO Models (SDXS-512-0.9): These models leverage the OpenVINO toolkit for accelerated inference on Intel CPUs. LCM-LoRA: This is a special type of model that can be used to fine-tune existing models or adapt them to specific styles. You can find more information about the supported models and how to use them on the FastSD CPU GitHub page: https://github.com/rupeshs/fastsdcpu

The developers of FastSD CPU are actively working on adding new features, so it's possible that custom checkpoint support might be added in the future. You can keep an eye on the project's updates on GitHub for any announcements.

Please note that loading custom checkpoint models might require additional technical knowledge and modifications to the software. If you are not familiar with these processes, it is recommended to wait for official support or seek assistance from the FastSD CPU community.

[AlgorithmicKing]

anyways thanks a lot

[rupeshs]

FastSD supports diffuser model format (safetensors) not single file checkpoint

[AlgorithmicKing]

ok