File: comfyui_00257_.png (1.71 MB, 1344x1024)

1.71 MB PNG

ah, my eyes

File: 00053-849332481705422-kro(...).jpg (2.55 MB, 3072x2304)

2.55 MB JPG

>>110117014
>ah, my eyes

File: 00054-786689331334703-kro(...).jpg (1.56 MB, 3072x2304)

1.56 MB JPG

>>110117090
daphne anon (pbuh) would have enjoyed that one

File: 1792464806162.jpg (229 KB, 768x1152)

229 KB JPG

>>110117109
Can't believe it's only been 3 years since I wanted to put her on a wolverine costume

File: 1792464848773.jpg (190 KB, 768x1152)

190 KB JPG

>>110117109
This one too

File: 00055-256921758892981-kro(...).jpg (1.8 MB, 3072x2304)

1.8 MB JPG

>>110117116
good things come to those who wait

>mfw Resource news

10/19/2026

>Kromoss: Full fine-tune of MOSS-Image-Base at 1024 px
https://huggingface.co/lodestones/Kromoss

>kromoss_int8_convrot.safetensors
https://huggingface.co/Clybius/Kromoss-Quantizations/tree/main

>Video Diffusion Avatars: Taming Video Diffusion Models into Controllable 3D Avatars
https://simongiebenhain.github.io/video-diffusion-avatars

>A Language-Informed Loss for Visual Generation
https://suchiragarwal.com/lilie

>Xiaomi-Lens-0 Technical Report
https://seerray-lab.github.io/Xiaomi-Lens-0

>ComfyUI v0.42.1
https://github.com/Comfy-Org/ComfyUI/releases/tag/v0.42.1

>Soran’t: Web front end for making video and music on your own ComfyUI
https://github.com/pwillia7/sorant

>UniForge3D — Windows and 8 GB VRAM
https://github.com/HardeNNN/UniForge3D-Low-VRAM

10/18/2026

>MOSS-Image-SFT-4-steps v2
https://huggingface.co/akhaliq/MOSS-Image-SFT-4-steps

>FastSDCPU release v1.0.0-beta.600 with pixel space image generation
https://github.com/rupeshs/fastsdcpu/releases/tag/v1.0.0-beta.600

>Fizgig v8.6.0: improved MOSS-Image likeness
https://github.com/shootthesound/Fizgig/releases/tag/v8.6.0

>ComfyUI MOSS Prompt Enhancer
https://github.com/slmonker/ComfyUI-MOSS-PromptEnhancer

>UniForge3D: Joint 3D Reconstruction and Generation in a Shared Anchor-Guided Space
https://xishuxishu.github.io/uniforge3d-project

10/17/2026

>MOSS-Image-Base, 8-step data-space DMD
https://huggingface.co/lodestones/MOSS-Image-Base-8-steps-dmd

>MOSS-Image-SFT (INT8 ConvRot) for ComfyUI
https://huggingface.co/Abiray/MOSS-Image-SFT-INT8-convrot

>Run Models Optimally On Your Local NVIDIA AI System
https://nvidia.github.io/local-ai-inference-recipes

10/16/2026

>MOSS-Image: Text-to-Image Generation with Native Layout Control
https://openmoss.ai/MOSS-Image

>MOSS-image ComfyUI Repackage
https://huggingface.co/Comfy-Org/MOSS-Image

>MiniMax-H3 Sol Engine in ComfyUI
https://github.com/NVlabs/Sana/tree/sol-engine/models/minimax_h3/Sol-H3-ComfyUI

>mfw Research news

10/19/2026

>MOSS-Image Technical Report
https://github.com/OpenMOSS/MOSS-Image

>Sink-Sparse Hybrid Attention for Efficient Inference in T2I/T2V Diffusion Transformers
https://arxiv.org/abs/2610.22975

>PreciseEdit: Strong Yet Stable Semantic Transport for One-Step Image Editing
https://arxiv.org/abs/2610.23535

>Generative AI Produces and Transmits Beauty Bias
https://arxiv.org/abs/2610.22553

>Rethinking the design space of Vanilla Pixel Diffusion Transformers
https://soumik-kanad.github.io/mflow-diffusion

>Kingfisher: Fast Personalized Visual Autoregressive Generation via Scale-Aware Prompt-Conditioned Scale Decomposition
https://arxiv.org/abs/2610.23238

>I2V-DETACH: Source Grounding Detachment for Unauthorized Image-to-Video Generation
https://arxiv.org/abs/2610.23287

>ProactGen: Learning Proactive Clarification for Video Generation
https://arxiv.org/abs/2610.22608

>FlashControl: One-Step Controllable Generator via Distillation-Friendly Single-Stream Teachers
https://arxiv.org/abs/2610.22537

>SCALE: Improving Guidance-Free Visual Generation via Self-Contrastive Alignment for Likelihood Estimation
https://jiwanhur.github.io/SCALE

>FedRAPP: Federated Test-Time Prompt Tuning for Vision-Language Models with Reliability-Aware Prompt Personalization
https://arxiv.org/abs/2610.23705

>PONDER: Learning to Generate in the Right Order
https://arxiv.org/abs/2610.23531

>SemPurify: Semantics-Aware Data Purification for Black-box Backdoor Defense
https://arxiv.org/abs/2610.23589

>MTEdit: Packed Trajectories Beat Rollouts for Multi-Turn Image Editing
https://arxiv.org/abs/2610.23751

>OmniXRef: A Unified Framework for Multimodal Multi-Reference Video Editing
https://oliver-cong02.github.io/OmniXRef.github.io

>Can We Use Saliency Maps for the Evaluation of Generative Image Models?
https://arxiv.org/abs/2610.23411

>Agentic Image Generation in the Wild
https://arxiv.org/abs/2610.23269

File: 00057-649774282768810-kro(...).jpg (1.45 MB, 3072x2304)

1.45 MB JPG

time to crash
gn all

File: debo_t_dms_00118_.jpg (646 KB, 2176x1408)

646 KB JPG

>>110117150
gn

File: 🐸00000_81118_.png (2.74 MB, 913x1229)

2.74 MB PNG

>>110117150
Gn

File: 1792466912902.jpg (203 KB, 768x1152)

203 KB JPG

Now it looks like an Alex Ross drawing

File: 1792467013610.jpg (232 KB, 768x1152)

232 KB JPG

>>110117150
Gn anon.

File: debo_t_dms_00120_.png (3.67 MB, 2176x1408)

3.67 MB PNG

>>110117272
nice

File: 1792474130820.jpg (275 KB, 1152x768)

275 KB JPG

Nailed it!

i miss schizo anon

File: 03738-captcha-cfg=1.0-ste(...).jpg (425 KB, 1200x1800)

425 KB JPG

gm

>gm

File: comfyui_00263_.png (2.83 MB, 1536x1024)

2.83 MB PNG

>>110119493
good morning

File: debo_rr_dms_00001_.png (1.36 MB, 2048x1280)

1.36 MB PNG

gm
another ldg melty?

File: 00000-757889326411254-kro(...).jpg (2.06 MB, 2304x3072)

2.06 MB JPG

kromoss (11k steps which i think is final)
shit is whack

File: debo_rr_dms_00002_.png (1.46 MB, 2048x1280)

1.46 MB PNG

>>110121152
RIP

File: 00002-817383604602259-kro(...).jpg (1.9 MB, 3072x2304)

1.9 MB JPG

>>110121205
that means crazy good

File: debo_rr_dms_00003_.png (897 KB, 2048x1280)

897 KB PNG

>>110121260
oh.. yippee, then

File: 00003-492884840959273-kro(...).jpg (2.46 MB, 3072x2304)

2.46 MB JPG

>>110121285
you better believe it
he also put out the dmd2 8step lora which is better than the other dmd lora because reasons

File: debo_rr_dms_00004_.png (2.65 MB, 2048x1280)

2.65 MB PNG

>>110121298
would you be down to do some side-by-sides?

File: 00006-810000901879596-kro(...).jpg (2.23 MB, 3072x2304)

2.23 MB JPG

>>110121439
comparing what?

File: debo_rr_dms_00005_.png (3.58 MB, 2048x1280)

3.58 MB PNG

>>110121546
moss to kromoss to kromoss dmd

File: out.jpg (2.22 MB, 6656x1248)

2.22 MB JPG

>>110121555
moss 40 step, kromoss 40 step, moss+dmd2 10 step, kromoss+dmd2 10 step
seeds2 simple, cfg 1 (cfg 3 without dmd2)
pos
>iphone photo, casual photograph of a redhead wearing a summer dress, drinking a tall glass of tea, in the french riviera, watching crowds, best quality
neg
>bad quality, distorted, deformed
https://files.catbox.moe/gyjebs.png
catbox is the kromoss dmd2

File: debo_rr_dms_00007_.jpg (1.15 MB, 2048x1280)

1.15 MB JPG

>>110121729
interesting. dmd seems to cause a bit of a 'flattening', for lack of a better word

>>110121463
This is cool

File: 00011-134861795957473-kro(...).jpg (2.14 MB, 3072x2304)

2.14 MB JPG

>>110121776
>flattening
of the boobs?
i would say dmd makes it a bit simpler, but it really depends on other settings (esp quality/style words) too. it does do what it's meant to do (make it 8 step) without losing quality tho. also one test is a very small sample lel
the benefit of kromoss is the additional knowledge, and of course you dont have to just use 8 steps or whatever with dmd

File: debo_rr_dms_00008_.jpg (991 KB, 2048x1280)

991 KB JPG

>>110121891
>i would say dmd makes it a bit simpler,
yea thats maybe more of what I meant. what stood out most to me was the uniform umbrella patterns

>>110121463
Looks like shit

File: 00013-343949700456004-kro(...).jpg (2.11 MB, 3072x2304)

2.11 MB JPG

>>110121986
maybe a seed change would help, although moss in general seems to like patterns

File: comfyui_00278_.png (2.59 MB, 1536x1536)

2.59 MB PNG

Surreal analog dreamscape merging Romantic nocturne painting, vintage space-age illustration, austere brutalist photography, candid naturalism, and psychedelic colored-pencil art. Monumental forms, severe geometric planes, strange rounded volumes; intimate tactile detail suspended within immense, silent depth. Deep Prussian blue, petroleum teal, moss green, weathered turquoise, concrete gray; concentrated eruptions of incandescent amber, acid yellow, vermilion, and electric cobalt. Dense blue-black shadows pierced by milky daylight, submerged illumination, and thin metallic glints. Light appears trapped inside translucent surfaces; reflected color trembles across surrounding darkness. Humid atmospheric veils, luminous haze, velvety distance, sharp foreground textures dissolving into soft obscurity. Scumbled oil pigment, translucent painted glazes, powdery pencil shading, granular printed color, restrained airbrush gradients; worn plaster, porous mineral surfaces, fibrous organic detail, tarnished metal. Rounded, almost rubbery contours meet uncompromising architectural edges. Precisely observed imperfections coexist with impossible scale and quietly irrational spatial relationships. Broad areas of stillness interrupted by intricate miniature detail and pockets of exuberant chromatic absurdity. Uneasy tenderness, deadpan strangeness, secluded grandeur, faded futuristic optimism; an abandoned utopia remembered through an old illustrated book. Unified material presence, deliberate asymmetry, strong silhouettes, deep tonal structure, selective saturation, palpable silence.

File: debo_rr_dms_00010_.png (3.52 MB, 2048x1280)

3.52 MB PNG

>>110122070
I yoinked a bunch of your prompts and turned them into one big randomizer

File: comfyui_00279_.png (2.02 MB, 1344x1344)

2.02 MB PNG

>>110122078
nice. they are probably over-worded, but oh well

File: K2-00083-sleek hard-surfa(...).png (1.25 MB, 768x1152)

1.25 MB PNG

Morning anons

File: debo_rr_dms_00015_.png (3.44 MB, 2048x1280)

3.44 MB PNG

>>110122163
the more words, the better

>>110122180
gm

File: 00014-400554934455489-kro(...).jpg (2.47 MB, 3072x2304)

2.47 MB JPG

>>110122180
gm

File: 00016-771357687503138-kro(...).jpg (1.74 MB, 2304x3072)

1.74 MB JPG

>cfg 10
ridin' that train, high on cocaine

>>110122208
>the more words, the better
this but unironically

File: comfyui_00280_.png (3.14 MB, 1344x1344)

3.14 MB PNG

>>110122240
and how many steps are you using

File: 00017-771357687503138-kro(...).jpg (1.92 MB, 2304x3072)

1.92 MB JPG

>>110122253
12 and dmd2 lora

File: 00018-771357687503138-kro(...).jpg (1.95 MB, 2304x3072)

1.95 MB JPG

cfg 10 may be a bit much lel

File: comfyui_00281_.png (1.49 MB, 512x1536)

1.49 MB PNG

>>110122294
i learned the hard way

File: 00019-771357687503138-kro(...).jpg (2.2 MB, 2304x3072)

2.2 MB JPG

cfg 8

File: comfyui_00283_.png (2.21 MB, 1536x1024)

2.21 MB PNG

Atmospheric surrealism fusing weathered industrial retrofuturism, antique fresco, nocturnal silver-gelatin photography, rain-soaked tonal painting, and feverish scratchboard expressionism. Sculptural rounded masses, recessed circular details, measured mechanical seams; ornate curling contours interrupted by blunt utilitarian geometry. Tangible miniature craftsmanship, painted enamel, rubbed edges, minute fastenings, accumulated grime. Chalky plaster, faded mineral pigments, tarnished gilding, veined stone, damp concrete; surfaces carrying distinct histories of abrasion, moisture, and touch. Dense charcoal blacks, pewter gray, dusty blue, muted ivory, smoky umber, deep bottle green; sparse, forceful accents of furnace orange and ember red. Broad diagonal sunlight rakes across uneven surfaces, exposing every ridge; delicate branching shadows fracture pale planes. Thin electric filaments flare through darkness, their reflections bending across curved glossy surfaces. Silvery halation, luminous rain streaks, trembling reflected silhouettes, humid distance. Photographic solidity threaded with obsessive hand-drawn marks: wiry crosshatching, scratched white contours, irregular stippling, concentric bands, rhythmic dashes disappearing into shadow. Intricate passages gathered around large, quiet shapes; repeated verticals disrupted by warped curves and sudden oblique cuts. Selective sharpness, soft optical falloff, fine film grain interwoven with visible pigment and paper tooth. Restrained elegance undercut by a tense, almost animate strangeness. Stillness charged with latent motion, melancholy grandeur, private unease, ordinary materials made quietly uncanny. Coherent light and perspective bind the antique, mechanical, organic, and feverishly graphic into one physical world.

File: 00022-877332625678044-kro(...).jpg (2.6 MB, 3072x2304)

2.6 MB JPG

>Checkpoint: update 11000 of the fine-tune (EMA weights), 704k training images.
this dude trained kromoss in what, 2-3 days lel

File: comfyui_00284_.png (2.7 MB, 1536x1024)

2.7 MB PNG

>>110122371
and negative:

Anime, manga, chibi, cartoon mascot, cute character design, flat vector art, cel shading, generic fantasy illustration, generic cyberpunk, steampunk clichés, decorative gears, excessive machinery, indiscriminate ornament, glossy showroom render, pristine plastic, sterile surfaces, chrome everywhere, uniform roughness, procedural grunge, artificial distress overlays, oversaturated rainbow palette, neon wash, candy colors, heavy orange-and-teal grading, flat illumination, excessive bloom, starburst filters, lens flare, HDR halos, oversharpening, crunchy clarity, featureless blacks, muddy midtones, uniformly blurred detail, fake depth of field, chaotic scribbles obscuring form, disconnected collage fragments, competing light sources, arbitrary perspective, visual clutter, theatrical gore, stock horror imagery, exaggerated expressions, sentimental nostalgia, captions, watermarks, logos.

File: 00026-10301577957814-krom(...).jpg (3.06 MB, 3072x2304)

3.06 MB JPG

>>110122437
nice

>>110122314
This is pretty cool

File: 00028-695203798970067-kro(...).jpg (1.68 MB, 3072x2304)

1.68 MB JPG

>>110122543
thx, sometimes it just works

File: comfyui_00289_.png (3.28 MB, 2048x1152)

3.28 MB PNG

Hallucinatory mixed-media surrealism fusing Renaissance elegance, misregistered pop-art silkscreen, tactile Gothic miniature photography, underground horror engraving, and drifting multiple-exposure imagery. Graceful serpentine contours buckle into jagged ornamental forms; classical proportions acquire strange asymmetries, serrated edges, and densely inscribed surfaces. Sculptural weight beneath flat printed color. Vermilion ink, apricot peach, pale cyan, sulfur yellow, chalk white; submerged layers of moss green, weathered cream, midnight violet, and cobalt blue. Broad matte color fields interrupted by bruised photographic shadows and needle-bright highlights. Thin yellow contour tracings, cyan registration offsets, ragged white reserves, coarse halftone screens, uneven ink coverage, exposed paper grain. Intricate black penwork cuts through luminous haze: wiry hatching, stippled cavities, tiny glyphlike incisions, nervous ornamental scratches. Rough mineral crust, aged bronze, leathery folds, polished glass; material detail rendered with the physical conviction of a photographed practical-effects model. Repeated translucent outlines drift out of alignment, multiplying verticals and bending perspective into a wavering architectural afterimage. Recognizable forms persist inside the distortion. Deep spatial recession meets abrupt graphic flatness; tightly cropped foreground fragments open onto distant, dissolving layers. Selective macro focus, razor-thin planes of clarity, liquid reflections, broad violet blur, cold blue rim light. Hard theatrical illumination catches brittle ridges while softer passages sink into opalescent darkness. Antique beauty, lurid pulp menace, hypnotic optical intimacy; ceremonial stillness disturbed by a slow visual tremor. Controlled clashes of scale, surface, and medium, held together by recurring curves and concentrated color.

What's the best solution for perfect media to photoreal pipeline? I've been trying all sorts of techniques, loras, models, controlnets, since before edit models became a thing. Img2img can get very realistic outputs at higher denoising but it hallucinates details like outfit, hairstyle, etc. edit models got the closest to perfect solution, since they can transfer the outfits, hairstyle, pose, background perfectly, but then they fall into 2 areas:
- outputs a real person with some aspects transfered from the reference, but changed details/hallucinations.
- outputs a perfect 1:1 with realistic materials and textures, but keep bad stylized anatomy, huge heads, etc.

In an ideal world, I'd like a model good enough to understand that I want to essentially recreate the reference AS a real picture, not a cosplay with a wig, not a 1:1 realistic person with anime proportions, but a "what would this picture look like if it was a real life photo, live action still" kind of thing.

Currently, the closest I can get is krea edit with a bunch of loras, but since editing with krea is a gimmick, it's still not perfect. Previously the best solution was Klein, but Klein fell strictly into those 2 problems I mentioned. One or the other.
I know saas models can do a very good job by default, but the filter and censorship makes it unusable. Just wondering if someone out there is at this same battle and found some solution for these problems.

File: 00041-459640289312297-kro(...).jpg (2.18 MB, 3072x2304)

2.18 MB JPG

>>110123553
did you try klein or qwen image 2.1, possibly with identiy and photoreal loras

File: comfyui_00294_.png (2.58 MB, 1936x1088)

2.58 MB PNG

An improbable collision of late-Shōwa illustrated pulp, nineteenth-century Romantic landscape painting, antique Japanese silk-screen naturalism, weather-beaten documentary photography, and obsolete scientific picture books. Dreamlike pictorial reality, alternately painstaking and crudely reproduced; the hushed precision of hand-painted botanical specimens interrupted by theatrical silhouettes, lurid printed fragments, strange optical intrusions, and inexplicable modern artifacts. Delicately modeled organic forms, fine insectile detail, brittle architectural geometry, velvety foliage, smoky atmospheric recession, immense distances swallowed by luminous milky fog. Monumental darkness beside featherweight translucency; awkward photographic literalism beside exquisite decorative invention. Flattened Japanese pictorial space abruptly opening into immense European landscape depth. Figures and forms suspended between convincing physical existence and the artificial stillness of an illustrated fable.

A peculiar chromatic instability: oxidized peach, tobacco brown, faded vermilion, dusty mauve, parchment ivory, moss green, bruised indigo, soot black, and pale mineral blue; occasional violent incursions of phosphorescent chartreuse, electric violet, sulfur yellow, and radioactive emerald. Thin golden contours emerging from muddy black masses. Bleached daylight, lavender twilight, dim interiors, and improbable fluorescent illumination coexisting without explanation. Deep shadow retaining a granular, almost fibrous density; distant forms melting into pearly atmospheric haze.

first half of the overly-long chatgpt prompt

File: comfyui_00295_.png (1.86 MB, 1344x1024)

1.86 MB PNG

second half:

Dry pigment dragged across absorbent paper, translucent mineral washes, delicate stippling, hand-retouched photographic surfaces, uneven lithographic inks, aged albumen-print tonality, soft emulsion bloom, accidental color misregistration, faint halftone contamination, scratched negatives, dust embedded in varnish, yellowed paper edges, and the fragile chromatic decay of a forgotten printed reproduction. Selective sharpness giving way to vague, bruised passages of near-abstraction.

Uneasy shifts of scale and perspective, oversized empty expanses, abruptly cropped forms, strange separations of foreground and distance, austere architectural masses, meticulous peripheral ornament, enigmatic graphic interruptions. The visual psychology of obsolete educational illustrations, supernatural magazine covers, neglected family albums, and museum paintings remembered imperfectly. Tenderness without sentimentality; beauty touched by abandonment, pastoral serenity contaminated by quiet absurdity. Unforced, elusive, slightly ridiculous, faintly ominous; a world simultaneously ancient, mundane, and inexplicably otherworldly.

File: debo_rr_dms_00027_.png (2.05 MB, 2048x1280)

2.05 MB PNG

>>110123926
cozy

>>110123926
>>110123999
really shitty style

File: debo_rr_dms_00028_.png (3.89 MB, 2048x1280)

3.89 MB PNG

>>110124099
did you accidentally look in a mirror?
fufufufuf gottem

File: ComfyUI_MOSS_00477_.png (1.04 MB, 1024x1024)

1.04 MB PNG

Trained a Ken Currie lora, here's a very cute migu :)

File: comfyui_00305_.png (1.89 MB, 1344x1024)

1.89 MB PNG

https://files.catbox.moe/e4zw2q.png

File: 00044-30533908137519-krom(...).jpg (2.43 MB, 3072x2304)

2.43 MB JPG

>>110125036
>dat prompt
mah nigga

i want to live in a creepy manor

>>110126171
Thank you for this nigbobump i was really worried about /sdg/ falling off the catalogue again

File: moss_00056_.png (2.9 MB, 1936x1088)

2.9 MB PNG

we back

File: K2-00084-steampunk robot (...).png (1.26 MB, 768x1152)

1.26 MB PNG

Afternoon/Evening anons
Good to see the site is back

>ga

File: 00001-416448746839008-kro(...).jpg (2.49 MB, 2304x3072)

2.49 MB JPG

>>110126280
howdy

So, what's the best current model for anime gen?

>>110126548
just use chatgpt desu

File: 🐸00000_81140.png (2.03 MB, 1264x800)

2.03 MB PNG

>Too many models.. Can'..keep.up

>>110126559
I just want to UOOOOH

File: 00004-344744682709305-kro(...).jpg (2.46 MB, 3072x2304)

2.46 MB JPG

at first i said "what the fuck" but it slowly grew on me

File: 🐸00000_81163.png (2.13 MB, 1200x848)

2.13 MB PNG

>>110126950
Tons going on, love the airships.

File: 00007-1107842706401884-kr(...).jpg (1.97 MB, 3072x2304)

1.97 MB JPG

>>110126995
indeed

File: debo_rr_dms_00032_.png (3.17 MB, 2048x1280)

3.17 MB PNG

oh cool, we're back
4chud needs to vibe code themselves a new server already

File: 01260-captcha-cfg=1.0-ste(...).jpg (450 KB, 1800x1200)

450 KB JPG

How about that. PHP was fixed.
>>110127047
Might need to try it out.

File: debo_rr_dms_00037_.jpg (1.08 MB, 2048x1280)

1.08 MB JPG

>>110127650
>PHP was fixed.
for now

File: 00005-87210693489962-krom(...).jpg (1.75 MB, 3072x2304)

1.75 MB JPG

>>110127650
>>110127661
some days it's nice to be a lowly no one in the chain of responsibility at work
>oh shit broke? not my problem, wake me up when it's fixed

File: 01380-captcha-cfg=1.0-ste(...).jpg (458 KB, 1200x1800)

458 KB JPG

gn anon

File: 00006-621672984203613-kro(...).jpg (1.87 MB, 3072x2304)

1.87 MB JPG

>>110127906
gn

File: 00019-435263537392810-kro(...).jpg (3.86 MB, 3072x2304)

3.86 MB JPG

gn all

File: debo_rr_dms_00038_.jpg (905 KB, 2048x1280)

905 KB JPG

>>110127906
>>110128344
gn

File: tmp-1792646899479.jpg (135 KB, 910x1338)

135 KB JPG

>>110128787

File: debo_rr_dms_00045_.png (2.95 MB, 2048x1280)

2.95 MB PNG

>>110129361
its art. you wouldnrt get it

My PC can't run Minimax. Is there NovelAI like site, but for Minimax / video generation?

i miss schizo anon

File: 02546-captcha-cfg=1.0-ste(...).jpg (539 KB, 1200x1800)

539 KB JPG

gm

new
>>110131788
>>110131788