Sign In
homemodelsimagesvideos3D Modelsarticlescomicsbountieschallengesupdatesshop

Flux1 Dev GGUF 8GB VRAM ControlNet Upscaler/Refiner. - v1.0 Showcase

NE

newbprofi

Loading Images

refiner

8gb

controlnet

flux

upscaler

Flux ControlNet image upscaler (with minor refine) with tweaked params for minor image changes/corrupt, by default for 8GB VRAM, but also can be modified for non-GGUF flux models.

Workflow: Input image / Output image.

Models:

flux dev gguf: https://huggingface.co/city96/FLUX.1-dev-gguf/resolve/main/flux1-dev-Q4_K_S.gguf

flux controlnet: https://civitai.com/api/download/models/901095?type=Model&format=SafeTensor

flux vae: https://huggingface.co/f5aiteam/VAE/resolve/main/ae.safetensors

clip 1: https://huggingface.co/f5aiteam/CLIP/resolve/main/clip_l.safetensors

clip 2: https://huggingface.co/city96/t5-v1_1-xxl-encoder-gguf/resolve/main/t5-v1_1-xxl-encoder-Q8_0.gguf

solf upscaler: https://huggingface.co/Acly/Omni-SR/blob/main/OmniSR_X4_DIV2K.safetensors

Notes/guide:

By default configured for upscaling/finetuning normal quality images to 1080p resolution. If you need upscale very bad/corrupted image, you may turn-on AUX node (also requires disconnect VAE Encoder and connect Empty LATENT to sampler). Also contains disabled by default Detail version of sampler, you need to disable default KSampler if you want try Detail one, and reconnect output image to VAE decode.

Deis sampler was chosen because in my opinion it gives less polished results, with more realistic imperfections and skin (compared to euler or dpmpp_2k).

Potentially it may be enhanced by faces/hands enhancing workflows (with florance2) and become even better.

Comments