topogoogles/topogoogles5

Fine-tuned FLUX model for generating photorealistic, cinematic portraits of the topogoogles5 subject.

Public
143 runs
Weights
Playground API Examples README Versions

Examples

View more

Run time and cost

This model costs approximately $0.0015 to run on Replicate, or 666 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia H100 GPU hardware. Predictions typically complete within 1 seconds.

Readme

topogoogles5

A fine-tuned FLUX image-generation model for creating photorealistic, cinematic images featuring the trained topogoogles5 subject.

Trigger word

Include topogoogles5 in every prompt to activate the trained subject.

Getting started

Describe the subject first, then add clothing, pose, location, lighting, camera style, and mood.

Example prompt

topogoogles5, cinematic editorial portrait, wearing a light turquoise tank top, neutral gray studio backdrop, soft rim lighting, sharp focus on the eyes, natural skin texture, shot on Hasselblad X2D 100C with an 85mm f/1.4 lens, shallow depth of field, photorealistic

Prompt ideas

Studio portrait

topogoogles5, high-end fashion studio portrait, black tailored blazer, softbox lighting, dark charcoal backdrop, confident expression, medium-format photography, detailed skin texture, editorial photography

Cinematic scene

topogoogles5 walking through a neon-lit city street after rain, night scene, wet pavement reflections, atmospheric fog, cinematic color grading, 35mm film look, dramatic side lighting

Lifestyle image

topogoogles5 relaxing in a modern Barcelona apartment, warm morning window light, casual linen shirt, candid lifestyle photography, natural colors, shallow depth of field

  • Model: dev for best quality; schnell for rapid iterations
  • Inference steps: around 28 with dev
  • Guidance scale: try 2 to 3.5 for natural-looking images
  • LoRA scale: start at 1.0; lower it for a subtler identity effect
  • Aspect ratio: choose based on composition, such as 1:1 for portraits, 4:5 for social posts, or 16:9 for cinematic scenes
  • Seed: reuse a seed when you want reproducible results

Image-to-image and inpainting

You can upload an input image for image-to-image generation. Adjust prompt_strength to control how much the result departs from the source image:

  • 0.2–0.5: Preserve more of the original composition and structure
  • 0.6–0.8: Balanced transformation
  • 0.9–1.0: Stronger prompt-driven transformation

Upload a mask as well to use inpainting for targeted edits.

Responsible use

Use this model only for lawful, respectful, and consensual creative work. Do not generate deceptive content, impersonate people without consent, or create harmful, exploitative, or privacy-violating imagery.

Model created