search

invoke-ai / InvokeAI

InvokeAI is a leading creative engine for Stable Diffusion models, empowering professionals, artists, and enthusiasts to generate and create visual media using the latest AI-driven technologies. The solution offers an industry leading WebUI, supports terminal use through a CLI, and serves as the foundation for multiple commercial products.
https://invoke-ai.github.io/InvokeAI/
Apache License 2.0
23.05k stars 2.38k forks source link

[bug]: Model Manager can not import LoRa made by FluxGym #6916

Open Barafu opened 6 days ago

Barafu commented 6 days ago

Is there an existing issue for this problem?

Operating system

Windows

GPU vendor

Nvidia (CUDA)

GPU model

No response

GPU VRAM

No response

Version number

5.0.0

Browser

Brave

Python dependencies

No response

What happened

I crerated a LoRa using FlexGym, carefully leaving all settings on default, except for epochs. When importing this LoRa with model manager, it produces the error.

[2024-09-23 12:28:05,823]::[ModelInstallService]::ERROR --> Model install error: D:/AI/webui_forge/webui/models/Lora/blue-ball.safetensors
InvalidModelConfigException: Unknown LoRA type: D:\AI\webui_forge\webui\models\Lora\blue-ball.safetensors

What you expected to happen

Lora be imported.

How to reproduce the problem

LoRa here: https://zippyshare.day/0kDGvKITOanFlBt/file Download and try to import.

Additional context

This LoRa works successfully in Forge. Has to set mode manually to "bnb-nf4 (fp16 LoRa)". And in Comfy.

Discord username

barafu_albino

Barafu commented 5 days ago

Still affects 5.0.0