Sign In

Beauty Enhancer + Realistic eyes

120.9k

Updated: Sep 14, 2026

conceptmalebeautywomaneyesbeautiful

Download

1 variant available

fp16 SafeTensor

Beauty_Eyes_Enhancer_flux_dev_3323751_epoch_10.safetensors

Half precision, best balance • 327.88 MB

Verified:

Type
LoRA
Stats

32

44

89

Reviews
Published

Sep 14, 2026

Base Model

Flux.1 D

Training
Steps: 4,000
Epochs: 10
Usage Tips
Strength: 1
Hash
AutoV2
EAD08A89A6
default creator card background decoration
Followers - 3087

3.1K

Likes - 14781

14.8K

The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.

IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

AH83RYCT64WEMB4S20KY97PRD0.jpeg

This lora makes people look more beuatiful/handsome while creating realistic eyes.

PONY + Flux v1.0

Although Flux is very good at creating perfect eyes, there is still a difference in sharpness and details when using this lora:

Strength:

  • use CFG 7 and no need for big strength (you can start from 0.2 and explore)

  • see the example photos for good options

  • positive and negative both can be used

  • Pony likes Euler_a based on my experiments with a few models

  • Flux has an extremely strong effect on face/body shape, so make sure to explore very small numbers, even as small as 0.025 (!), etc.

SDXL v1.0

Here is a comparison of how some of the most used eye enhancing loras execute the same prompt:

Same prompt with Beauty Enhancer + Realistic eyes lora:

I'll let you decide which you like most: -)

Strength:

  • seems to work well between 0.50 to 1.85 but you may use negative strength as well (see example in the gallery)

  • positive strength might bring the person a bit closer and create a more simplistic background while enhancing realism

Negative strength:

  • Negative can be used too

  • negative might distance the person a bit and create a richer background while losing from realism

  • it has a special drawing/cartoon effect around -1.00 which needs the expression 'realistic eyes' in the prompt (othervise in general 'realistic eyes' is NOT needed)

Colors the lora was trained on:

  • amber, blue, brown, green, gray, hazel

Have fun exploring it and please post your results here!