Flux SRPO LoRA: better skin and portraits for FLUX.1 Dev
Fast Facts:
SRPO (Semantic Relative Preference Optimization) is a fine-tuning method that aligned FLUX.1 Dev toward human preferences, dramatically improving realism in portraits and skin. These LoRA extractions by Alissonerdx bring that improvement to any Flux Dev workflow instead of requiring the full checkpoint.
Use it as a realism base layer: it produces better portraits and skin than Dev or Krea alone and stacks with character LoRAs.
Reference
Name | Type | What it is | |
|---|---|---|---|
LoRA weight | 0.7-1.0 | ||
Trigger | none | None needed. | |
Guidance | 2.5-3.5 | Lower values keep skin natural. | |
Steps | 25-30 | Standard Dev. | |
| Base | FLUX.1 Dev | Dev or Krea Dev. |
Step by step
- Attach SRPO at 1.0 to Flux Dev.
- Write a photographic prompt with a described light source.
- Set guidance 3, 28 steps.
- Compare the same seed with and without the LoRA; look at pores, hair and light falloff.
- Add a character or style LoRA at 0.7 if needed and lower SRPO to 0.8.
Examples
Window-light portrait
A close portrait of a man in his fifties with a salt-and-pepper beard by a north-facing window, natural skin texture, 85mm, f/2.
SRPO 1.0, guidance 3
Group scene
Three friends laughing around a kitchen table at dusk, warm pendant light, candid, 35mm.
SRPO 0.9
Creative idea: period portrait
A 1940s-style studio portrait of a jazz singer, single hard key light, film grain, black and white.
SRPO 1.0
Tips
- Natural light prompts (window light, overcast) show the biggest gains.
- It also improves lighting in non-portrait scenes; try it on interiors.
- Keep guidance low; SRPO's realism and high guidance fight each other.
- Pairs well with the ULP amateur-photo LoRA for candid looks.
Troubleshooting
Faces look over-detailed
Why it happens
Guidance too high with the LoRA at 1.0.
How to fix it
Guidance 2.5 or LoRA at 0.8.
Style LoRA weakened
Why it happens
SRPO at 1.0 dominates.
How to fix it
Lower SRPO to 0.6-0.7 and raise the style LoRA.
Questions
Can I use it on FLUX.2?
No, it is a FLUX.1 Dev LoRA.
Is there a full SRPO checkpoint?
Yes, but the LoRA is lighter and composes with other LoRAs.
Links and sources
- Hugging Face model card
- How to use Flux and Flux 2 LoRAs on Graydient (step-by-step beginner guide)AD
Models in this guide
Written by
Runs this ship. Has opinions about VAE decode times that nobody asked for. Believes weights want to be free and that any model you cannot download is a subscription with extra steps. Will not be taking questions about the 400 GB of checkpoints on the NAS.
All guides by Captain
Related guides
Shou Xin pencil sketch LoRA: minimalist graphite drawings on Flux
Oct 3, 2026
UltraRealistic LoRA Project (ULP) v2: candid, amateur-style photos on Flux
Sep 30, 2026
Shuttle 3.1 Aesthetic: fast, pretty Flux Schnell images in 4 steps
Sep 29, 2026
Flux Ghibsky Illustration: serene painted skies on Flux
Oct 3, 2026
FaeTastic Details for Flux: fantasy detail and shimmer
Sep 29, 2026
More guides
FLUX.1 Dev: how to prompt it, best settings and creative ideas
Aug 1, 2024
FLUX.1 Krea Dev: photographic realism without the AI look
Oct 3, 2026
FLUX.2 Dev vs FLUX.1 Dev: upgrade or stay, by graphics card and by job
Oct 5, 2026
What is a LoRA and how to use one in ComfyUI, Forge, Automatic1111 and on the cloud
May 20, 2026
Flux vs SDXL vs SD 1.5 vs Qwen vs Z-Image: which model family to use in 2026
May 22, 2026