Sign In

METAFILM 元影 SD3.5m FP16 Asian realistic ★ 亚洲摄影风格 DPM++2M

19
294
2
13
Verified:
SafeTensor
Type
LoRA
Stats
294
2
Reviews
Published
Nov 3, 2024
Base Model
SD 3.5 Medium
Training
Epochs: 10
Usage Tips
Strength: 0.8
Hash
AutoV2
8E6BF923E9
This Stability AI Model is licensed under the Stability AI Community License, Copyright (c) Stability AI Ltd. All Rights Reserved.
Powered by Stability AI
Please support civitai and creators by disabling adblock

★元影★-SD3.5m-★METAFILM-AiARTiST FP16★亚洲摄影 LoRA★

🔥 全新的图像生成模型 SD3.5m FP16 摄影风格LoRA

🔥 A brand new T2i model SD3.5m high-quality realistic photo

----------------------------------------------------------------------

CFG推荐4.5,大家可以在3~5之间自行调节。

DPM++2M sgm_uniform采样步数默认22,可酌情调节

可以不加载T5xxl,显存占用7.9GB,速度和画质比较平衡

CFG recommends 4.5, and adjust it between 3-5 .

sampling steps for DPM++2M sgm_uniform are 22,

which can be adjusted as needed

T5xxl is not necessary, VRAM occupies 7.9GB,

the speed and image quality are relatively balanced

----------------------------------------------------------------------

————————————————————————————————————————

测试问题请留言,业务合作看个人首页 +V Zyuan980

AiARTiST | Metafilm 中国·山东 数字人/AiGC/元宇宙方向

做好工具人 服务艺术家 联系方式:https://x1f3ewlrcf.feishu.cn/wiki/BjJ1waQaLitPB4k7Lbvc0MaVnzb?fromScene=spaceOverview&open_tab_from=wiki_home

————————————————————————————————————————

Discussion

MysticMindAi's Avatar
MysticMindAi

What is the reason for this lora? Is it so we don't have to use the T5?