import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("OpenMOSS-Team/MOSS-Image-Base", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("aoleg/MOSS-Image-Kodachrome")
prompt = "Kodachrome film, middle-aged white man with a weathered face wearing black overalls, gloves and goggles on his forehead. He stands beside a large wooden Northern Pacific Railway freight car, holding a coiled hose in his right hand. His left hand rests on the car's door. The background shows a clear blue sky and railway tracks. , 1940s, photo by Jack Delano, 4x5 Kodachrome"
image = pipe(prompt).images[0]MOSS-Image Kodachrome
- Prompt
- Kodachrome film, middle-aged white man with a weathered face wearing black overalls, gloves and goggles on his forehead. He stands beside a large wooden Northern Pacific Railway freight car, holding a coiled hose in his right hand. His left hand rests on the car's door. The background shows a clear blue sky and railway tracks. , 1940s, photo by Jack Delano, 4x5 Kodachrome
- Prompt
- Kodachrome film, middle-aged white man with a receding hairline wearing a green headband, blue denim jacket and brown pants. He stands with arms crossed, smiling slightly, in an industrial yard with railroad tracks and smokestacks in the background. A yellow circular object is on the ground near his feet. , 1940s, photo by Jack Delano, 4x5 Kodachrome
- Prompt
- Kodachrome film, muscular middle-aged man with short, graying hair and a beard wearing a blue, sleeveless, knitted vest over a white t-shirt, washing his hands at a rusty outdoor sink. Behind him, two other men are partially visible. A blue can with white sailboat graphics is on the sink's edge. Dark warehouse, 1940s, photo by Jack Delano, 4x5 Kodachrome
- Prompt
- Kodachrome film, grizzled, middle-aged man coal miner with dirty, sweat-streaked face wearing dark green overalls, gloves and a cap with goggles. He has a cigarette in his mouth and is leaning on a soot-covered machine in a dimly lit, industrial setting. The background is dark and cluttered with metal, 1940s, photo by Jack Delano, 4x5 Kodachrome
- Prompt
- Kodachrome film, middle-aged man with dirty, weathered skin, blue eyes and a slight smile. He wears a dark dirt-stained flat cap and an olive-green, button-up shirt. His face and cap are heavily smeared with grime, suggesting manual labor. The background is dark and blurred focusing attention, 1940s, photo by Jack Delano, 4x5 Kodachrome
- Prompt
- Kodachrome film, Aerial photo of a snowy train yard with multiple train cars lined up on parallel tracks. Thick white smoke billows from several locomotives, creating a misty atmosphere. The ground is covered in a thin layer of snow and the sky is clear blue. Train cars are stacked on either side of the tracks, casting long shadows. , 1940s, photo by Jack Delano, 4x5 Kodachrome
- Prompt
- Kodachrome film, middle-aged white man wearing a green cap, glasses, blue patterned sweater and gray pants, operating a vintage train control panel inside a dimly lit train cab. He has a short beard and is focused on the controls. The background shows large windows with a view of an outdoor train station and overcast, 1940s, photo by Jack Delano, 4x5 Kodachrome
- Prompt
- Kodachrome film, focused, bearded man with dark hair wearing a light brown shirt and dark vest, sitting at a cluttered wooden desk in a dimly lit, rustic office. He's writing in an open notebook under a black, industrial-style lamp. Papers, maps and notes are pinned and scattered around the wooden-paneled walls. , 1940s, photo by Jack Delano, 4x5 Kodachrome
- Prompt
- Kodachrome film, bustling train yard featuring a bright orange and yellow locomotive at the center, surrounded by multiple tracks. People in winter clothing are scattered around, some near a red tanker car on the left. Industrial buildings and smokestacks are visible in the background under an overcast sky. The ground is a mix of gravel and dirt, 1940s, photo by Jack Delano, 4x5 Kodachrome
- Prompt
- Kodachrome film, industrial steel mill interior. Three men in grey work clothes and black hats stand around a large green, steam-powered hammering machine with a glowing orange steel bar being struck. The machine is centered surrounded by tools and metal rods on the floor. Large, grid-patterned windows in the background let in natural light, illuminating, 1940s, photo by Jack Delano, 4x5 Kodachrome
Model description
base_model: OpenMOSS-Team/MOSS-Image-Base tags: - lokr - lycoris - moss-image - style - kodachrome - film - ai-toolkit license: apache-2.0
Kodachrome LoKr for MOSS-Image
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, but above all the quality and the sharpness. It was trained on MOSS-Image-Base with Ostris AI Toolkit at 1280 batch 4 for 750 steps; the style had transferred by then, but not yet the grain. Treat it 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. Trigger words: `Kodachrome film, Kodachrome`. Describe the scene in the prompt as you normally would; the film look comes from the adapter.
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: 750, about 3.5 epochs
- EMA: on, decay 0.98
- Timestep sampling: flowmatch, shift, balanced
- Resolution: 1280 (area cap), bucketing on, no square crop
- Precision: bf16
- Checkpoints every 250 steps; the 750 step checkpoint is the one 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.
Trigger words
You should use Kodachrome film or Kodachrome to trigger the image generation, but rather use something like shot on Kodachrome film or the trigger may bleed into the image (see the front of the diesel engine).
Download model
Download them in the Files & versions tab.
- Downloads last month
- 59
Model tree for aoleg/MOSS-Image-Kodachrome
Base model
OpenMOSS-Team/MOSS-Image-Base