-
Tested with Pony Diffusion, I used both a Controllite SDXL Canny Model and a Controlnet SDXL line art model, neither of them were able to generate. I am using the latest version of the repo.
Exampl…
-
I'm trying to use a quantized [GGUF model](https://huggingface.co/city96/FLUX.1-dev-gguf). I've set up the models as per https://github.com/lllyasviel/stable-diffusion-webui-forge/discussions/1050.
…
-
While the other models work just fine (flux1-dev-fp8/flux1-dev-bnb-nf4-v2) generating images with flux1-dev result in a random color image.
While it generate the image i can see normal results, but a…
-
The full error log is:
To load target model KModel
To load target model ControlNet
Begin to load 2 models
[Memory Management] Current Free GPU Memory: 6295.45 MB
[Memory Management] Required Mo…
-
-
It might be an upstream issue.
I'm using Forge with default settings, except for the resolution. However, I’ve tried most of the samplers and schedulers to fix the problem, but without success. Wha…
-
**Describe the bug**
在我使用simulator进行验证的时候，它输出了nan。我使用yolov8的onnx以及kmodel去验证我的kmodel_input.bin和onnx_input.bin余弦相似度为0.99，但是一切换到yolov5的onnx和kmodel，他就变成nan了
**Origin model and code**
这是生成kmodel的代码
`…
-
Does anyone know why I'm getting OOM?
Flux 8bit, 512x512, 20 steps. Forge. I created a Lora myself, doesn't work. I created that lora again via citivai and captioned it much better, Nonetype error.…
-
![image](https://user-images.githubusercontent.com/1320252/57378873-3d7d3080-71d8-11e9-9f…
-
Help me figure out how to switch models via api? If I need to support both sd and flux via api.
The problem is that the api performance depends on what I manually enable, what checkpoints I switch …