flux-photo - v6.0
Highlighted images
Recommended Prompts
An old man in a wooden fishing boat
Recommended Parameters
samplers
steps
cfg
Tips
Use transparent alpha or beta for a more precise effect and finer details with less impact on composition; apply weight from 0.5 to 3.0 like a slider.
Use v6.0 for stronger, impressive photo and detail enhancement; start with LoRA strength of 1.0 but experiment between 0.2 and 2.0 as needed.
Keep prompts simple when using v6.0, and only add realistic tags if results are unsatisfactory.
v5 is trained to improve performance at smaller steps using a fine-tune approach.
V4-compatible versions can be used for standard, extreme, or nuanced effects by adjusting weights accordingly.
V3-high-strength works well without any realism prompts; V3-medium-strength complements prompts already containing realism tags.
Please use transparent alpha or beta for a more precise effect that won't change the overall composition too much. Use v6.0 for a stronger impressive photo/detail/realism enhancement.
Transparent Alpha and Beta: (more compatible, finer details)
Trained to target finer details only
More compatible
Preserves Context Better
Use like a slider, weight 0.5-3.0
V6: (stronger effect, can affect context)
Use simple prompts - "An old man in a wooden fishing boat",
Use Lora strength 1.0, but experiment with 0.2 - 2.0 as needed. Has a wide weight range
Only add 'realistic' or 'fine hairs, visible pores, etc.' if you're not getting results
dies sampler+simple scheduler or Euler sampler+beta/simple scheduler (others may also work well)
Guidance 2 - 5 (usually 3.5)
Steps 20 a 50 (usually 25 or 35)
Improved dataset, diffusers training back end used via SimpleTuner, weight tuned with custom script.
V5:
Improved dataset, trained for flux-dev on the flux-dev2pro fine-tune designed to improve training by pushing the model to work with smaller steps.
***Note: In my demo photos I had the file named 'flux-photo_v6.0_turbo' but it's the same file, there is no v6_turbo. I went through several versions and canned all but this one that stood out. Turbo? You might get away with a few less steps, but I didn't want to put turbo in the name.
V4-compatible/standard/extreme:
standard is usually the best, extreme with weight scaled down can be nice for exaggerated or forced photo realism, but can eat styles and turn things into normal people. compatible is pretty nuanced, could be nice with a higher weight.
V3-high-strength:
excellent on it's own with no prompt input about realism
V3-medium-strength:
excellent added to prompts that already have other loras or language about photorealism so it doesn't overbake the generation
flux lora trained on high quality photography images to produce realistic high quality photos without needing to add any tags or keywords (i.e. no 'a photo of' or 'realistic' etc.)
Contributor
Model Details
Discussion
Please log in to leave a comment.
