|
--- |
|
tags: |
|
- text-to-image |
|
- flux |
|
- lora |
|
- diffusers |
|
- template:sd-lora |
|
- fluxgym |
|
widget: |
|
- output: |
|
url: sample/ldlaughingmemeface_000450_00_20241001223402.png |
|
text: LDME Young adult male at a party, mid-laugh with eyebrows raised in mock |
|
surprise. He's holding a red solo cup and wearing a graphic t-shirt |
|
- output: |
|
url: sample/ldlaughingmemeface_000450_01_20241001223612.png |
|
text: LDME Close-up of a smirking businessman with arched eyebrows, squinting |
|
eyes, and a slightly open mouth. He's holding a coffee mug in an office setting |
|
- output: |
|
url: sample/ldlaughingmemeface_000450_02_20241001223822.png |
|
text: LDME middle-aged woman in a casual shirt, raising his eyebrows skeptically |
|
while holding a martini |
|
base_model: black-forest-labs/FLUX.1-dev |
|
instance_prompt: LDME |
|
license: other |
|
license_name: flux-1-dev-non-commercial-license |
|
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md |
|
--- |
|
|
|
# LDLaughingMemeFace |
|
|
|
A Flux LoRA trained on a local computer with [Fluxgym](https://github.com/cocktailpeanut/fluxgym) |
|
|
|
<Gallery /> |
|
|
|
## Trigger words |
|
|
|
You should use `LDME` to trigger the image generation. |
|
|
|
How to Use This LoRA Model |
|
|
|
Describing the Expression |
|
|
|
To get the best results from this LoRA model, use descriptive words that capture the essence of the facial expressions you want to generate. Consider phrases like: |
|
|
|
"exaggerated expression" |
|
"raised eyebrows" |
|
"as if mocking" |
|
"eyebrows lifted and smirking" |
|
|
|
These descriptors will help the model focus on creating the unique, expressive faces you're aiming for. |
|
|
|
Adjusting LoRA Strength |
|
|
|
Strength 1: At full strength, the LoRA model tends to generate images of the same character when you prompt with "a man." This can be useful if you want consistency in character appearance. |
|
|
|
Dialing Down Strength: If you wish to create different characters or add variety, reduce the strength of the LoRA. A strength setting around 0.7 has been effective for me, but I encourage experimenting with different levels to achieve your desired outcome. |
|
|
|
## Download model and use it with ComfyUI, AUTOMATIC1111, SD.Next, Invoke AI, Forge, etc. |
|
|
|
Weights for this model are available in Safetensors format. |
|
|
|
|
