README.md

---
tags:
  - flux
  - image-to-image
  - lora
  - diffusers
  - fal
base_model: black-forest-labs/FLUX.1-Kontext-dev
instance_prompt: Generate youtube thumbnails using text "Your text"
license: other
license_name: flux1-dev-non-commercial-license
license_link: >-
  https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev/blob/main/LICENSE.md
widget:
  - text: Generate youtube thumbnails using text "EPIC FAIL."
    output:
      url: images/1.png
  - src: inputs/1.png
    prompt: Generate youtube thumbnails using text "EPIC FAIL."
    output:
      url: images/1.png
  - text: Generate youtube thumbnails using text "MIND BLOWN!"
    output:
      url: images/2.png
  - src: inputs/2.png
    prompt: Generate youtube thumbnails using text "MIND BLOWN!"
    output:
      url: images/2.png
  - text: Generate youtube thumbnails using text "CAN'T BELIEVE IT!"
    output:
      url: images/3.png
  - src: inputs/3.png
    prompt: Generate youtube thumbnails using text "CAN'T BELIEVE IT!"
    output:
      url: images/3.png
  - text: Generate youtube thumbnails using text "fal.ai is best.". Hands around his head.
    output:
      url: images/4.png
  - src: inputs/4.jpeg
    prompt: Generate youtube thumbnails using text "fal.ai is best.". Hands around his head.
    output:
      url: images/4.png
---
# Youtube Thumbnails Kontext Dev LoRA

<Gallery />


## Model description

Works best at resolution 1536x1024 input images (works on other resolutions as well)

Recommended lora scale: 0.4-0.5

## Trigger phrase

Generate youtube thumbnails using text "Your text"

## Download model

Weights for this model are available in Safetensors format.

[Download](https://huggingface.co/fal/Youtube-Thumbnails-Kontext-Dev-LoRA/resolve/main/thumbnails_lora_rank_32.safetensors)

## Training at fal.ai

Training was done using [fal.ai/models/fal-ai/flux-kontext-trainer/playground](https://fal.ai/models/fal-ai/flux-kontext-trainer/playground).