DFL多角度素材生成工具

特困生

meiyouruguo 发表于 2026-2-28 20:53
偶尔会发生, 模型有时候不识别提示词

好的，收到

coolkyanyi

存符石，能貸款嗎？哈哈

woshihaha

jandown 发表于 2026-1-13 10:15
建议加个香蕉的api，比开源的qwen-edit 效果要好很多

香蕉收费啊

z445726864

零食不够，发帖慢慢攒

o4andv1

最低多少G显存用

meiyouruguo

o4andv1 发表于 2026-3-16 17:14
最低多少G显存用

12G勉强可以,  16G推荐

z445726864

meiyouruguo 发表于 2026-3-16 17:25
12G勉强可以,  16G推荐

我5070  12G  49张差不多个把小时。。效果不错

o4andv1

Using pytorch attention in VAE
Using pytorch attention in VAE
VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16
Requested to load WanVAE
loaded completely; 7449.00 MB usable, 242.03 MB loaded, full load: True
Found quantization metadata version 1
Using MixedPrecisionOps for text encoder
CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16
Requested to load QwenImageTEModel_
loaded completely; 9581.80 MB usable, 7910.28 MB loaded, full load: True
Requested to load WanVAE
Unloaded partially: 3633.28 MB freed, 4277.00 MB remains loaded, 453.25 MB buffer reserved, lowvram patches: 0
loaded completely; 3104.87 MB usable, 242.03 MB loaded, full load: True
loaded completely; 9320.65 MB usable, 7910.28 MB loaded, full load: True
model weight dtype torch.float8_e4m3fn, manual cast: torch.bfloat16
model_type FLUX
got prompt


处理出错

ryan1944

很棒的素材生成工具

mukdswzdkum

处理出错,  恳请大佬帮解~感谢您