aoleg/MOSS-Image-Kodachrome-V2

MOSS Image Kodachrome V2

Prompt

Kodachrome film, two man railroad workers in early 20th-century attire, standing beside a large black steam locomotive. One wears a red cap and black jacket, the other a gray shirt and overalls, holding a fire extinguisher. Background includes industrial buildings and railway tracks under a cloudy sky. , 1940s, 4x5 Kodachrome

Prompt

Kodachrome film, three man railroad workers in dark blue overalls and caps, climbing and working on the side of a large black steam locomotive. One worker is standing on top, while the other two are mid-climb. The background shows a cloudy sky and a tall utility pole with multiple wires. Railroad tracks and mechanical parts are visible, 1940s, 4x5 Kodachrome

Prompt

Kodachrome film, male pilot and brown hair wearing a brown leather aviator cap, beige flight suit and black boots, climbing into a weathered silver WW II-era military aircraft with a yellow stripe. He has a beige backpack slung over his shoulder. The background shows a clear blue sky with wispy clouds. , 1940s, 4x5 Kodachrome

Prompt

Kodachrome film, man and woman military personnel in beige uniforms and caps. Woman, standing on the ground, assists the man, who is seated on a ladder attached to a blue military aircraft. The seated man wears brown shoes and looks down at the assisting woman. Background includes the aircraft's wing and part of a hangar. Bright daylight, 1940s, 4x5 Kodachrome

Prompt

Kodachrome film, woman machinist and curly brown hair wearing a brown mesh cap, blue short-sleeve shirt and black apron. He is focused on operating a metal lathe machine, holding a wooden tool in his right hand and adjusting the machine with his left. His expression is concentrated. , 1940s, 4x5 Kodachrome

Prompt

Kodachrome film, snowy mountainous landscape featuring a cluster of white and beige buildings with snow-covered roofs. The buildings are surrounded by snowdrifts and scattered rocks. In the background, snow-capped mountains rise under a partly cloudy blue sky. A small tree stands near the buildings and a winding path cuts through the snow in the foreground. , 1940s, 4x5 Kodachrome

Prompt

Kodachrome film, misty overcast forest landscape. Snow-dusted evergreen trees dominate the midground with a large fog-shrouded mountain in the background. The foreground features a grassy meadow with patches of snow. Colors are muted with greens, browns and whites, creating a serene, wintery atmosphere. , 1940s, 4x5 Kodachrome

Prompt

Kodachrome film, woman with short brown hair wearing a black dress, sitting on the floor against a colorful, modular wall display. The wall features geometric panels in yellow, orange, pink, blue and green with mannequin heads and an eye panel. Background includes checkered drawers and patterned curtains. Floor is tiled in gray. , 1950s

Prompt

Kodachrome film, modern, indoor-outdoor lounge area with wooden ceiling and floor, large glass windows and black structural beams. Two people, one in an orange shirt and the other in a red shirt, sit at a table. Colorful circular cushions and potted plants adorn the space. Sunlight filters through trees outside, creating a bright, 1950s

Prompt

Kodachrome film, six smiling women in 1950s attire, posing energetically on and around a wooden table in an elegant room. They wear checkered and pastel dresses with one woman in a blue shirt. Arms raised, they exude joy. Background features ornate wall sconces and double doors. The women's hairstyles are

Prompt

Kodachrome film, Walt Disney, a middle-aged Walt Disney with slicked-back brown hair and a smile wearing a dark gray suit, white shirt and blue tie. He stands behind a large detailed globe showing North and South America. On top of the globe is a plush toy of Mickey Mouse, dressed in red shorts and yellow shoes, 1950s

Prompt

Kodachrome film, group of eight adults, six men and two women, sitting around a small campfire on a rocky lakeshore. They wear yellow rain ponchos and outdoor gear with one man in a blue sweater kneeling to tend the fire. The background features a calm lake, large rocks and a forested hill under a cloudy sky, 1950s

Prompt

Kodachrome film, snowy rural scene featuring a mint-green and white vintage car loaded with luggage, two brown dogs standing in front, a man in dark clothes standing to the left and a woman in a brown coat and hat near the car's open trunk. The white wooden house with red shutters and snow-covered roof is in the background, 1950s, 35mm Kodachrome

Prompt

Kodachrome film, sunny day on a sailboat. Five people, three men and two women, stand on the deck. The central man with short brown hair, wears a red polo and yellow pants, holding a brown folder. To his right, a blonde woman in a pink shirt and floral skirt holds binoculars wearing sunglasses. , 1960s, 35mm Kodachrome

Prompt

Kodachrome film, two women in a red speedboat on a sunny day. The woman on the left has light brown hair, wears a pink headscarf, black and white striped tank top and black shorts with an American flag to her left. She has a light tan and smiles at the woman beside her, who has blonde hair covered, 1960s, 35mm Kodachrome

Prompt

Kodachrome film, vintage living room with floral wallpaper, wooden blinds and mid-20th century decor. Features a dark wooden TV cabinet with an old CRT TV, a green air conditioner, striped armchair, floor lamp with beige shade, framed pictures on the wall, landscape painting, small side table with white lampshade, carpeted floor, 1950s

Prompt

Kodachrome film, woman with freckles wearing a yellow headscarf, aviator goggles and a dirty, buttoned-up work shirt with a dark cardigan. She holds a wooden shovel, standing beside a large black steam locomotive with visible steam. In the background, there's a blurred dark, 1940s, 4x5 Kodachrome

---

license: apache-2.0 tags: - lokr - lora - text-to-image - style - kodachrome - film

Kodachrome LoKr for MOSS-Image V2

This is a LoKr adapter that gives MOSS-Image the look of Kodachrome slide film. It mimics the color response and the contrast curve of 1940s 4x5 Kodachrome sheet film, capturing its quality and its sharpness. It was trained on MOSS-Image-Base with Ostris AI Toolkit at 1280 batch 4 for 1000 steps at low noise timestep bias.

V2 vs. V1

The first version, MOSS-Image-Kodachrome, was trained with the Balanced preset for timesteps. I found that this made the LoKr leak some composition properties from the dataset. I retrained it with mostly the same parameters, changing only to Low Noise timestep bias. After 1000 steps, the fine detail of the 4x5 sheet film, its tonal curve and its color balance had been learned, without the composition bleed.

V2 is more subtle than V1: no composition bleed, it applies mostly the colors and the tonal curve, and it accentuates fine detail. It is not necessarily better than V1, only different.

V1 was trained for 750 steps, about 3.5 epochs. The style had transferred by then, but not yet the grain, so treat V1 as an intermediate checkpoint.

Usage

Load it with the standard LoRA loader in ComfyUI (LoKr support is in core) or with load_lora_weights in Diffusers. Trained and tested on MOSS-Image-Base; tested at strength 1.0 with CFG 3, 50 steps, and skip-layer guidance on layer 22. Describe the scene in the prompt as you normally would; the film look comes from the adapter.

Trigger words

Use Kodachrome film or Kodachrome. A phrase like shot on Kodachrome film works better; a bare trigger may bleed into the image (see the front of the diesel engine).

Why a factor 16 LoKr

A film look is a small amount of information: a palette, a tone curve, grain. A full-rank LoKr at factor 16 has enough capacity for that and produces a file about 5% the size of a rank 32 LoRA. In community Krea 2 and MOSS runs, LoKr also shows less subject bleed than LoRA, which matters for a style trained on photographs of people and places; the adapter should carry the film and not the dataset.

Dataset

551 scans of Kodachrome slides, high resolution, no JPEG compression artifacts. Subjects and settings are deliberately varied: portraits, street scenes, landscapes, interiors, vehicles, daylight and artificial light, several decades of emulsion.

Images were trained at native resolution with aspect-ratio buckets, capped at a 1280² pixel area (1.64 MP); scans larger than that were downscaled, smaller ones left as they are. The set bucketed into 130 resolution groups, 66 of them single images. Captions describe the scene content and carry the trigger word; the film look itself is not described in the captions.

Training

Ostris AI Toolkit, architecture moss_image (MOSS-Image-Base), text encoder frozen.

• Network: LoKr, full rank, factor 16
• Optimizer: AdamW 8-bit, learning rate 1e-4
• Batch size 4 (within buckets, so 213 steps per epoch; mean effective batch 2.6)
• Steps: 1000, about 4.7 epochs (V1: 750)
• EMA: on, decay 0.98
• Timestep sampling: flowmatch, shift, low noise bias (V1: balanced)
• Resolution: 1280 (area cap), bucketing on, no square crop
• Precision: bf16
• Checkpoints every 250 steps; the 1000 step checkpoint is published here

Limitations

Grain and halation are learned at 1280² area and weaken at output sizes well below 1 MP. Inherits MOSS-Image's known weak points (small text, hands). Tested on Base only; behavior on MOSS-Image-SFT is unverified.

License

The adapter is released under Apache-2.0, matching the base model. The training images are not included.

Download

Get the files from the Files & versions tab.