(HUNYUAN v1) Trigger word "Andrea Botez". Drag and drop a video for the workflow I used. Follow THIS GUIDE to train your own Hunyuan video LoRA on Windows. Love my models and want to support me? Send some buzz my way. If you're feeling extra generous and would like to donate to help support the costs of training these resource intensive models you can directly donate to me though HERE.
(FLUX v2) No trigger needed. Improved likeness with new training method and extensive captions. Trigger words are no longer needed. Just prompt what you want.
Love my models and want to support me? Send some buzz my way. If you're feeling extra generous and would like to donate to help support the costs of training these resource intensive models you can directly donate to me though here
(FLUX v1)For proper likeness you MUST FP16 clip. Tested and works with the fp8 Model as long as you are using the fp16 CLIP. Love my models and want to support me? Send some buzz my way. If you're feeling extra generous and would like to donate to help support the costs of training these resource intensive models you can directly donate to me though here
Description
FAQ
Comments (3)
you made the same mistake than me, you put rank 64 so 600+ mb file (I think yo could achieve same quality with rank 32, maybe a bit more epoch at best) overall very good likeness, but it's true chess board have 64 squares, so in way it make sense :)
Yep I used defaults for the first couple I trained. Future models will be Rank 32! ~300mb file
can you point me to a tutorial to train my own?
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.