Instructions to use multimodalart/tarot-ideogram-4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use multimodalart/tarot-ideogram-4 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("ideogram-ai/ideogram-4-nf4-diffusers", torch_dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("multimodalart/tarot-ideogram-4") prompt = "machine learning" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
| tags: | |
| - text-to-image | |
| - lora | |
| - diffusers | |
| - fal | |
| base_model: | |
| - ideogram-ai/ideogram-4-nf4-diffusers | |
| license: other | |
| widget: | |
| - text: machine learning | |
| output: | |
| url: https://cdn-uploads.huggingface.co/production/uploads/624bebf604abc7ebb01789af/z4OCxDSeJleXezS6P_NCU.png | |
| # tarot ideogram 4 | |
| <Gallery /> | |
| ## Model description | |
| A tarot card LoRA trained with the public domain card set style of Raider Waite 1920. [Dataset](https://huggingface.co/datasets/multimodalart/1920-raider-waite-tarot-public-domain) | |
| ## Trigger words | |
| This is a style LoRA, without a trigger word, trained with json captions | |
| Example: a card for `machine learning` | |
|  | |
| ```json | |
| { | |
| "high_level_description": "A robed figure stands upon a mountain of data descending a winding path, haloed by a glowing lattice of interconnected nodes while streams of numbers cascade from the sky.", | |
| "compositional_deconstruction": { | |
| "background": "A twilight landscape rendered in flat allegorical style. A deep indigo sky above gives way to a stepped mountain of stacked geometric strata below, suggesting layered terrain. Faint grid lines recede toward a low horizon. Cool palette of indigo, teal, and electric cyan with warm amber accents.", | |
| "elements": [ | |
| { | |
| "type": "obj", | |
| "bbox": [ | |
| 300, | |
| 360, | |
| 780, | |
| 660 | |
| ], | |
| "desc": "Central standing figure in a long flowing robe, one hand raised toward the glowing lattice above and the other gesturing down the path, face calm and forward-facing." | |
| }, | |
| { | |
| "type": "obj", | |
| "bbox": [ | |
| 120, | |
| 250, | |
| 420, | |
| 760 | |
| ], | |
| "desc": "A radiant halo of interconnected nodes arranged in three vertical layers behind the figure's head and shoulders, small circles joined by thin luminous lines forming a network." | |
| }, | |
| { | |
| "type": "obj", | |
| "bbox": [ | |
| 560, | |
| 200, | |
| 940, | |
| 820 | |
| ], | |
| "desc": "A stepped mountain of stacked horizontal strata descending from upper left to lower right, a narrow switchback path winding down its face toward a small glowing point at the base." | |
| }, | |
| { | |
| "type": "obj", | |
| "bbox": [ | |
| 60, | |
| 120, | |
| 380, | |
| 300 | |
| ], | |
| "desc": "Streams of falling numerals and ones-and-zeros cascading from the upper sky like rain on the left side, fading as they descend." | |
| }, | |
| { | |
| "type": "obj", | |
| "bbox": [ | |
| 420, | |
| 700, | |
| 560, | |
| 860 | |
| ], | |
| "desc": "A small glowing orb cradled in the figure's lowered hand, emitting concentric rings of light." | |
| }, | |
| { | |
| "type": "obj", | |
| "bbox": [ | |
| 820, | |
| 760, | |
| 940, | |
| 900 | |
| ], | |
| "desc": "A single bright marker at the foot of the path where the descent ends, suggesting a converged minimum." | |
| } | |
| ] | |
| } | |
| } | |
| ``` | |
| ## Download model | |
| Weights for this model are available in Safetensors format. | |
| [Download](/multimodalart/tarot-ideogram-4/tree/main) them in the Files & versions tab. | |
| ## Training at fal.ai | |
| Training was done using [fal.ai/models/ideogram/v4/trainer](https://fal.ai/models/ideogram/v4/trainer). |