HSPU = Has1.4-.35SXD8-.4proto5-.4uber
Description
# HSPU 1.1 = HassanBlend1.5 +
Has1-4--35SXDv8--4proto5 = HassanBlend1.4-.35SXDv8 + protogenX53OfficialR_10 + HassanBlend1.4-.35SXDv8, 0.4, ckpt
Has1-4--35SXD8--4proto5--4uber = O1 + uberRealisticDreamyP_uberRealisticDreamyP, 0.4, ckpt
FAQ
Comments (8)
Thank you - eager to try it out. But if you have an original PickleTensor version, could you upload it too? Safetensors currently can’t be used to fine-tune models without a conversion which may or may not lose information in the conversion.
v1.0 ckpt Add, do not have 1.1 ckpt use webui extension model converter.
Moreover fine-tune models can mix pre-existing concept were it can be better to merge by add difference for example A-HSPU B-myTrainingFaceSD1.5 C-SD1.5 at .95
@DigitalDreamer is the website deleted your new lora extract?
@1048926251350 no, only the image.
Can you add a description, recommended prompts, VAE to use, etc? Thanks.
You can see my prompts in "Generation Data" of the images... and in the info version i show the checkpoint recipe mix, used tags recommended by each model of the mix. I din't use any VEA, tbh model start to be old Source: trust me bro
What is this?
This is a checkpoint to generate an image from a prompt.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.