Instructions to use aoleg/MOSS-Sharpness-Slider with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use aoleg/MOSS-Sharpness-Slider with Diffusers:
pip install -U diffusers transformers accelerate
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-Sharpness-Slider") prompt = "A young woman in 1950s posing on a Cadillac, pinup style." image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
|
Download README.md from aoleg/MOSS-Sharpness-Slider: direct link, hf CLI and curl.
- Browser
- Download file 7.01 kB
-
https://huggingface.co/aoleg/MOSS-Sharpness-Slider/resolve/main/README.md
- Command line
-
hf download hf://aoleg/MOSS-Sharpness-Slider/README.md
-
curl -L -o README.md https://huggingface.co/aoleg/MOSS-Sharpness-Slider/resolve/main/README.md
7.01 kB
| tags: | |
| - text-to-image | |
| - lora | |
| - diffusers | |
| - template:diffusion-lora | |
| widget: | |
| - output: | |
| url: images/151716095-1636270155-moss_image_base_int8_convrot.png | |
| text: A young woman in 1950s posing on a Cadillac, pinup style. | |
| - output: | |
| url: images/151716096-1636270156-moss_image_base_int8_convrot.png | |
| text: A young woman in 1950s posing on a Cadillac, pinup style. | |
| - output: | |
| url: images/151716097-613179008-moss_image_base_int8_convrot.png | |
| text: >- | |
| Kodachrome film, man with short brown hair wearing a dark shirt and beige | |
| overalls, standing in a dimly lit workshop. He is meticulously working on | |
| the wooden nose of an antique, twin-engine aircraft with metal propellers. | |
| The aircraft's polished wooden surface contrasts with the dark background, | |
| highlighting the man's focused expression and skilled hands. , 1940s, B25 | |
| bomber, photo by Alfred Palmer | |
| - output: | |
| url: images/151716098-3872554109-moss_image_base_int8_convrot.png | |
| text: >- | |
| Kodachrome film, 1940s-style scene in a dimly lit workshop. A woman with | |
| curly brown hair wearing a red blouse, black skirt and red bandana, kneels | |
| beside a vintage olive-green truck. She holds a wrench and looks intently at | |
| the truck's window, where a man with short dark hair, 1940s | |
| - output: | |
| url: images/151716099-3300073256-moss_image_base_int8_convrot.png | |
| text: >- | |
| Kodachrome film, 1940s-style woman with dark hair in a pink headband wearing | |
| a white blouse, light blue capri pants and brown shoes with white socks. She | |
| sits against large silver, riveted aircraft fuel tanks, holding a red | |
| canister. An open black toolbox is on the ground to her right, 1940s, photo | |
| by Alfred Palmer | |
| - output: | |
| url: images/151716100-4081148363-moss_image_base_int8_convrot.png | |
| text: >- | |
| Kodachrome film, Polynesian-themed restaurant with a yellow awning | |
| displaying "POLYNESIAN RESTAURANT" in colorful letters. A tall, vertical | |
| sign with "POKE" stands to the left. Leafless tree in the center, people in | |
| winter clothes walking on the sidewalk and a streetlamp. Background includes | |
| palm trees, 1960s | |
| - output: | |
| url: images/151716101-1649273751-moss_image_base_int8_convrot.png | |
| text: >- | |
| Kodachrome film, smiling middle-aged white man with short dark hair wearing | |
| a light gray suit, white shirt and teal tie, holding a pointer stick. He | |
| stands to the left, pointing at a large detailed, colorful aerial map behind | |
| him, which depicts a park-like area with roads, buildings, green spaces, a | |
| lake, trees, 1950s | |
| - output: | |
| url: images/151716102-2245860517-moss_image_base_int8_convrot.png | |
| text: >- | |
| Kodachrome film, elderly white couple standing beside a vibrant red vintage | |
| convertible car with white-walled tires, parked on a gravel driveway in | |
| front of a grand, white-columned mansion. The woman with short gray hair, | |
| wears a navy blue sleeveless top and beige skirt, while the man, in a light | |
| blue short-sleeve shirt, 1950s, photo by Toni Frissell, 35mm Kodachrome | |
| - output: | |
| url: images/151716103-3684661186-moss_image_base_int8_convrot.png | |
| text: >- | |
| Kodachrome film, two young women in a sunlit, modern outdoor setting. The | |
| woman on the left has short blonde hair, wears sunglasses, a light beige | |
| cardigan over a blue romper and brown flats. She stands with hands in | |
| pockets, leaning slightly towards the other woman. On the right, the woman | |
| with red hair wears a white, 1950s, photo by Toni Frissell, 35mm Kodachrome | |
| - output: | |
| url: images/151716105-1869421269-moss_image_base_int8_convrot.png | |
| text: >- | |
| DSLR, two young women in a sunlit, modern outdoor setting. The woman on the | |
| left has short blonde hair, wears sunglasses, a light beige cardigan over a | |
| blue romper and brown flats. She stands with hands in pockets, leaning | |
| slightly towards the other woman. On the right, the woman with red hair | |
| wears a white, 2020s | |
| - output: | |
| url: images/151716106-891512518-moss_image_base_int8_convrot.png | |
| text: >- | |
| Kodachrome film, vintage 1950s-style street scene featuring a red and white | |
| striped "Lulu Belle" general store with ornate white trim. Two classic cars, | |
| a white and a silver, are parked in front. A black car with red taillights | |
| is partially visible in the foreground. The sky is overcast and the store, | |
| 1950s | |
| - output: | |
| url: images/151716107-4081738831-moss_image_base_int8_convrot.png | |
| text: >- | |
| Kodachrome film, 1940s-style nightclub scene. A blonde woman in a floral | |
| dress sings passionately into a vintage microphone, standing on a table. | |
| Seated around her are elegantly dressed men and women in formal attire, | |
| sipping drinks. The background features red velvet curtains and the lighting | |
| is warm, creating a nostalgic, glamorous atmosphere. , 1940s, photo by | |
| William Gottlieb, MF Kodachrome | |
| base_model: OpenMOSS-Team/MOSS-Image-Base | |
| instance_prompt: null | |
| license: apache-2.0 | |
| # MOSS Image Sharpness Slider | |
| <Gallery /> | |
| ## Model description | |
| # MOSS-Image Sharpness Slider | |
| A Concept Slider LoRA for MOSS-Image that moves an image between sharp and blurry by LoRA strength alone. Positive strength sharpens, negative strength blurs. No trigger word, no prompt changes. | |
| IMPORTANT: Intended usage is deblurring the MOSS Base checkpoint. It will have a weaker effect on MOSS SFT, and almost none on SFT+Turbo. The reason is that SFT is already tuned to filter out the blur, and Turbo distills it further. | |
| ## Usage | |
| Load as a regular LoRA on MOSS-Image Base or SFT. Start at 1.0 for sharper output and -1.0 for blur, then adjust. Strength 0 is the untouched model. No trigger word. | |
| ## Training | |
| Trained with the `concept_slider` process in Ostris AI Toolkit on MOSS-Image Base (int8 convrot). The slider direction comes from the prompt pair; the dataset only supplies latents at both ends of the range. | |
| - Dataset: 72 photos and the same 72 photos with Gaussian blur, 1024 px, uncaptioned. Default captions "sharp, in focus" and "blurry, out of focus". | |
| - Positive prompt: "a sharp photo, perfectly in focus, crisp fine detail, clear edges" | |
| - Negative prompt: "a blurry photo, blurred, out of focus, soft and smeared" | |
| - Target class and anchor: empty. Guidance strength 3. | |
| - LoRA rank 4, alpha 4. Layer 22 excluded from the LoRA; it is the skip-layer guidance block for MOSS, and `uncond_skip_layers: [22]` during training. | |
| - lr 0.001, adamw8bit, weight decay 0.001, EMA 0.99, batch 1, flowmatch with shift timesteps, bf16. | |
| - This is the step 400 checkpoint. Steps 350 and below had not converged; 400 gives the full effect without visible overfitting. | |
| ## Notes | |
| Each slider step runs several passes of the base model, so training is about 3x slower per image than a regular MOSS LoRA. 400 steps took well under two hours on a single consumer GPU. | |
| ## Download model | |
| [Download](/aoleg/MOSS-Sharpness-Slider/tree/main) them in the Files & versions tab. | |