Updated: May 17, 2024
base modelVerified: a year ago
SafeTensor
Type | |
Stats | 381 0 |
Reviews | (25) |
Published | May 2, 2024 |
Base Model | |
Hash | AutoV2 FFFAD2FB6B |
MAY OR MAY NOT WORK IN FORGE / A1111. I've only got it working in ComfyUI
Features:
Enhanced contrast
Enhanced colours
Enhanced blacks
Brighter whites
Very good intriciate details
A CosXL merge of dreamMODE. This will work in anything that supports v-prediction and Continuous EDM sampling.
Usage: DMP++ SDE sampler, karras, cfg 2-4. EDM Continuous sampling: v-prediction, sigma max 120, sigma min 0.002, try going up to 240 for sigma max.
NOTE: LORAs probably will not work.
Update: Some LoRAs will work, but may work in unexpected ways. Your mileage may vary.
Workflow:
See Image Edit Model Examples | ComfyUI_examples (comfyanonymous.github.io) for details.
This Stability AI Model is licensed under the Stability AI Non-Commercial Research Community License, Copyright (c) Stability AI Ltd. All Rights Reserved.