Instructions to use Rodrigo23Ramirez/yt-thumbnail-flux with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Rodrigo23Ramirez/yt-thumbnail-flux with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("Rodrigo23Ramirez/yt-thumbnail-flux") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
yt-thumbnail-flux
This model is a fine-tuned version of black-forest-labs/FLUX.1-dev trained with a custom LoRA to generate YouTube thumbnails with dynamic compositions and celebrity portraits.
Rodrigo Thumbnail Flux (LoRA + FLUX merge)
This model is a fine-tuned version of black-forest-labs/FLUX.1-dev using LoRA for YouTube thumbnail generation.
🧠 Use case
Generate dynamic, eye-catching YouTube thumbnails with celebrity faces, expressive poses, bold colors, and dramatic lighting.
📦 Training details
- Base model:
black-forest-labs/FLUX.1-dev - LoRA rank: 16
- Steps: 1500
- Resolution: 1024x1024
- Batch size: 1
- Optimizer: AdamW 8-bit
- Scheduler: FlowMatch
🧪 Example prompt
MYTHUMB style, YouTube thumbnail with two famous people, cinematic lighting, bold colors, dramatic expressions
- Downloads last month
- -
Model tree for Rodrigo23Ramirez/yt-thumbnail-flux
Base model
black-forest-labs/FLUX.1-dev