ユーザーLoRA
XL
PCM LoRAs of Stable Diffusion XL for Fast Image Generation
- アップロード日時
- 2024/08/25 18:40
- 採用数
- 2.0k
- レビュー
- レビューがありません
- 権限
- 画像生成と共有を許可する
- ユーザーによるモデルのダウンロードを許可する
- 生成画像の商用利用を許可する
ユーザーLoRA
Usage: LoRA Weight: 0.6 to 1.0. Sampling Steps: 6 to 7, higher is possible. Sampling Method: Euler a is safe. CFG Scale: 1.1 to 2.0 is safe, higher values may give poor results.
Note: It is generally best to avoid using this with models that already have Hyper SDXL merged, Lightning merged, or other turbo models.
Note: It seems to have a higher contrast than when using Hyper SDXL LoRA. If used with other LoRAs, the results may be different from what you expected.
https://github.com/G-U-N/Phased-Consistency-Model https://huggingface.co/wangfuyun/PCM_Weights https://civitai.com/models/487106/pcm-loras-of-stable-diffusion-xl-for-fast-image-generation