1 mb LORA trained in 5 mins that does the same thi
Back

1 mb LORA trained in 5 mins that does the same thi
0 0 0

Branding

1 mb LORA trained in 5 mins that does the same thi
1 mb LORA trained in 5 mins that does the same thi
1 mb LORA trained in 5 mins that does the same thi
1 mb LORA trained in 5 mins that does the same thi
1 mb LORA trained in 5 mins that does the same thi
1 mb LORA trained in 5 mins that does the same thi
1 mb LORA trained in 5 mins that does the same thi
1 mb LORA trained in 5 mins that does the same thi
1 mb LORA trained in 5 mins that does the same thi
1 mb LORA trained in 5 mins that does the same thi
I created this LORA to prove the point I made earlier. This LORA has the same functionality as the previously mentioned model but is improved because it is more portable, reduced in size by 1975 times, more versatile, does not require merging processing, and can easily adjust the intensity of the effects. The training process was completed on my phone, taking only 5 minutes and 10 seconds, with preparation taking another 5 minutes. I simply used six examples provided in the aforementioned model for training without any caption processing or other operations. I adopted the Luisap method mentioned in my previous post. Among the examples I included, some did not involve any prompt engineering from the original model, while others underwent engineering processing and used other LORAs. Please do not create checkpoints for highly specific concepts unless there is a very compelling reason. Additionally, when training LORAs for single characters, consider adding other characters or even styles to increase the utility of the LORA.
This model is sourced from an external transfer (transfer address: https://civitai.com/models/7525/1-mb-lora-trained-in-5-mins-that-does-the-same-thing-as-25-gb-model-but-better ）,if the original author has objections to this transfer, you can click，
Appeal
We will, within 24 hours, edit, delete, or transfer the model to the original author according to the original author's request

Sun

Sun

Branding

Model Information

Active
Original author：
Aili
Model Type：
LoRA
Basic Model：
SD 1.5
Resource Name：
doortoinfinity.safetensors
MD5：
ca519a13ed4009d5c9eff2fbec6a7f8c
I created this LORA to prove the point I made earlier. This LORA has the same functionality as the previously mentioned model but is improved because it is more portable, reduced in size by 1975 times, more versatile, does not require merging processing, and can easily adjust the intensity of the effects. The training process was completed on my phone, taking only 5 minutes and 10 seconds, with preparation taking another 5 minutes. I simply used six examples provided in the aforementioned model for training without any caption processing or other operations. I adopted the Luisap method mentioned in my previous post. Among the examples I included, some did not involve any prompt engineering from the original model, while others underwent engineering processing and used other LORAs. Please do not create checkpoints for highly specific concepts unless there is a very compelling reason. Additionally, when training LORAs for single characters, consider adding other characters or even styles to increase the utility of the LORA.
This model is sourced from an external transfer (transfer address: https://civitai.com/models/7525/1-mb-lora-trained-in-5-mins-that-does-the-same-thing-as-25-gb-model-but-better ）,if the original author has objections to this transfer, you can click，
Appeal
We will, within 24 hours, edit, delete, or transfer the model to the original author according to the original author's request