The LoRA model was built on 60 images taken from publicly available sources on the Internet. LoRA was trained on the FLUX Dev model
Improved image quality, works well with other FLUX models.
Weight: 0.3 - 1.
Works well with weight: (0.4 - 0.6), experiment with weight.
It depends on the model, but you can raise the Lora weight quite high.
High quality prompts reduce the negative effect.
Cinematic style as a separate art form is a unique combination of visual, audiovisual and narrative elements that together create a certain atmosphere and express the artistic vision of the director. The style is based on the use of certain techniques of camera work, editing, lighting, color, sound and acting.
Cinematic style can vary from realism to abstraction, from minimalism to an abundance of visual effects. Each film contributes to a certain style or genre, influencing the viewer's perception through emotional impact, artistic symbolism and cultural contexts. In this sense, cinematic style is a way of communicating and expressing ideas that goes beyond simple narrative, becoming a powerful tool of artistic expression.
Description
The LoRA model was built on 60 images taken from publicly available sources on the Internet. LoRA was trained on the FLUX Dev model
Improved image quality, works well with other FLUX models.
Weight: 0.3 - 1.
Works well with weight: (0.4 - 0.6), experiment with weight.
FAQ
Comments (6)
Great Lora, Good joB!
I'm curious about the size of V3 - it's over 1GB, which seems like a substantial jump from the 18mb of previous versions. I've noticed some quality improvements, but I'm wondering if they're worth almost 60 times increase in file size. Could you help me understand what's driving this increase and if it's really necessary to be that much larger?
its crazy why is it so much bigger?
larger network dim / alpha during LoRA training, usually means more training data details, nuances, are included; the fine aspects missed when doing lower network rates. And image fidelity, quality. With higher, if done right, it can learn the finer elements of art or imagery, such as lighting, shading, etc., the list goes on.
anybody know why v3 is so massive? v3 is 1gb . v2 is 18mb. fishy :/
larger network dim / alpha during LoRA training, usually means more training data details, nuances, are included; the fine aspects missed when doing lower network rates. And image fidelity, quality. With higher, if done right, it can learn the finer elements of art or imagery, such as lighting, shading, etc., the list goes on.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.