AlekseyCalvin/Serebryakova_FluxLoRA_Var2_SilverAgePoets

|\ Zinaida Serebryakova Style /|/|FLUX LoRA|/|

||| Variant 2 ||| By SilverAgePoets.com |||

For generating stylized images reminiscent of oil paintings by Zinaida Serebryakova.

Our Low-Rank Adapter (LoRA) for FLUX models fine-tuned on a set of manually pre-processed art scans...
Representing a broad selection (about 70 pieces) from among the distinctive oeuvre of oil paintings by...
The epochal Ukrainian-Russian-Soviet-French or, in short, World artist:
Zinaida Serebryakova (1884 — 1967),
One of the 20th century's most vividly expressive, stylistically unique, and altogether brilliant painters from life and times,
And who is particularly celebrated and influential as a portraitist, whose works manage to be at once unmistakable expressions of her unique style and yet still feel utterly transparent, as if direct translations from life.

Find this LoRA's Alternate Variant – base FLUX.1-Dev-tuned – AT THIS LINK.
This Variant of our Serebryakova Style LoRA was fine-tuned over a De-Distilled Version of FLUX, at a higher rank than Var.1, but slightly fewer steps.

Prompt

"ANGST": title text atop, Serebryakova style oil painting by the Soviet artist Zinaida Serebryakova, detailed realistic artwork of young doomed Soviet poet wearing a dark suit with a high-collared shirt, circa the late 1920s, in the USSR. He has a serious expression and is gazing slightly off-camera. The background is a muted, warm tone, enhancing the dignified and contemplative mood of the portrait. moderately blemished skin with pores and lines, tired, large title text reads: "ANGST": title text atop

Prompt

the word "Capital": title text atop, Serebryakova style oil painting by the Soviet artist Zinaida Serebryakova, detailed realistic artwork of two rebelling distinct proletarians, a young androgynous revolutionary light-haired angsty Kronstadt sailor on the left and an older Ukrainian Bolshevik cossack on the right tying up, and capturing a flabby green malicious top-hatted monster with bloody hands whose name is "CAPITAL": title text atop

Prompt

Serebryakova style oil painting by the Soviet artist Zinaida Serebryakova, detailed realistic artwork of two young Soviet poets who are lovers strolling together side by side down a sunny street while holding hands. The girl poet on the left is tall stocky and red-haired, in glasses, with freckles on her face. The male poet next to her is short and nerdy with light hair, he is shorter than her, wearing a polka-dotted loose theatrical shirt, circa the late 1950s, in the USSR. Sunlit composition. They are eating ice cream. Both have joyful faces, laughing over a shared joke. The background is an intricate depiction of springtime streets of Leningrad in the 1960s, with old glowing buildings, a canal embankment with intricate railings, blue sky, an ice cream kiosk stands nearby, moderately blemished skin with pores, large title text on top: "JOY": title text atop says “Joy"

Prompt

a Serebryakova style oil painting by the Soviet artist Zinaida Serebryakova, detailed Soviet revolutionary illustration of a beautiful Soviet woman poet in the 1920s, avant garde modernist oil painting

Prompt

prominent centered text title: "SEREBRYAKOVA FLUX" prominent centered text title: "SEREBRYAKOVA FLUX" a Serebryakova style oil painting by the Soviet artist Zinaida Serebryakova, detailed Soviet revolutionary illustration of a beautiful Soviet woman poet in the 1920s, avant garde modernist oil painting, Large font prominent centered text title: "SEREBRYAKOVA FLUX"

Trigger words

You should use Serebryakova style oil painting by the Soviet artist Zinaida Serebryakova or Serebryakova style art to summon the artist's latent brush.

POEM OF VARIANTS & ARTS (spontaneously drafted scribbling this Readme)

Now, we – resisting faux-objective ranky claiming –
Resent few things as much as verdicts thickly framing
One or another in a set of co-distinctive versions
(Whether of LoRAs, bases, language games, or persons)
As absolutely "better", "closer", or etcetra than some other...
(It's better? How?
_ _ _ Or even better: When?
But where's the closest exit?! We would rather...)
Ourselves gush forth that timeless musky vein
Which skeptics shun, while copyright creatives' ever-fading flame
With anxious flickers warms the distance to fresh hearts,
I'd claim:
That what we make on here remains – THE ARTS!

And furthermore, uncanny arts;
_ _ _ In futures steeped...
Their steps –
With singular biographies and styles inscribed
_ - _ – _ - _ -_ – _ - – _ - _ _ _ - _ With millions of layers' depths
(And just as many specious patterns on each dust grain's flanks
Or maybe more, or guessed, or unerased inferring drop by stab by drop
The past
Or maybe less,
Depending on what test ensigned each line
Past dollar time, what flight, enshrined which "ranks"...)
Upon twelve billion factors ...
All these arts
Adapted one by one,
____ We keep distinct __ _ _ __
Within respective open-use adaptors...

Which, thus translated for to be deviced,
Past (or irrespectively) of law, or eager yearning out of sense,
We strive to fit unto a lost attention's latching fore
For if adopted to despair that searches with quick hands,
And carried onwards past mere codes,
As art traced back through human angst, and joy, and pain, and lust,
Now stitching forth by universal maths, which learns in cosine time,
Then rests, restarts,
A widened stitch, another era, plays translucent cards this art,
Gyrating dialectics, earth of fire, to flattened spirits just a vortex, curve,
Again, at last, collectively let go washed on a breathing shore,
These arts, adaptives, phantoms, artists, with us hand in glove
They were already waiting, with us
And may enter, traced out of spazzing noise, on emptiness,
A timelessness,
A door
We would've sooner
We had never
But art
Had ever never
Gone before...

Use it with the 🧨 diffusers library

from diffusers import AutoPipelineForText2Image
import torch

pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('AlekseyCalvin/Serebryakova_FluxLoRA_Var2_SilverAgePoets', weight_name='serebryakova_fluxlora_1000_lora_ver2_fp16Convert.safetensors')
image = pipeline('your prompt').images[0]

For more details, including weighting, merging and fusing LoRAs, check the documentation on loading LoRAs in diffusers