novelstudio24/hiyala-ai 🖼️🔢❓📝✓ → 🖼️
About
Hiyala.ai brings the charm and elegance of Maldivian women and their traditional dress to life through AI-generated imagery.
Example Output
Prompt:
"Create a portrait of a young woman dhivehi hiyala standing in a rustic setting, framed by an open window made of natural materials like wood and thatched palm leaves. She is wearing a vibrant red traditional dress with intricate golden and multicolored embroidery around the neckline, accessorized with a long, gold necklace. Her expression is serene and confident, with her hands gently clasped in front of her. The background features lush green foliage, creating a natural and peaceful atmosphere. The lighting is soft and warm, emphasizing the details of her attire and the textures of the surrounding environment."
Output
Performance Metrics
10.07s
Prediction Time
10.16s
Total Time
All Input Parameters
{
"image": "https://replicate.delivery/pbxt/MIVXzTpj1nIJWNNvAUsgVIsTN0ZXyFWZTVm0j5iVt1YzZM1e/IMG_7538.jpeg",
"model": "dev",
"prompt": "Create a portrait of a young woman dhivehi hiyala standing in a rustic setting, framed by an open window made of natural materials like wood and thatched palm leaves. She is wearing a vibrant red traditional dress with intricate golden and multicolored embroidery around the neckline, accessorized with a long, gold necklace. Her expression is serene and confident, with her hands gently clasped in front of her. The background features lush green foliage, creating a natural and peaceful atmosphere. The lighting is soft and warm, emphasizing the details of her attire and the textures of the surrounding environment.",
"go_fast": false,
"lora_scale": 1,
"megapixels": "1",
"num_outputs": 1,
"aspect_ratio": "1:1",
"output_format": "webp",
"guidance_scale": 3,
"output_quality": 80,
"prompt_strength": 0.8,
"extra_lora_scale": 1,
"num_inference_steps": 28
}
Input Parameters
- mask
- Image mask for image inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
- seed
- Random seed. Set for reproducible generation
- image
- Input image for image to image or inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
- model
- Which model to run inference with. The dev model performs best with around 28 inference steps but the schnell model only needs 4 steps.
- width
- Width of generated image. Only works if `aspect_ratio` is set to custom. Will be rounded to nearest multiple of 16. Incompatible with fast generation
- height
- Height of generated image. Only works if `aspect_ratio` is set to custom. Will be rounded to nearest multiple of 16. Incompatible with fast generation
- prompt (required)
- Prompt for generated image. If you include the `trigger_word` used in the training process you are more likely to activate the trained object, style, or concept in the resulting image.
- go_fast
- Run faster predictions with model optimized for speed (currently fp8 quantized); disable to run in original bf16
- extra_lora
- Load LoRA weights. Supports Replicate models in the format <owner>/<username> or <owner>/<username>/<version>, HuggingFace URLs in the format huggingface.co/<owner>/<model-name>, CivitAI URLs in the format civitai.com/models/<id>[/<model-name>], or arbitrary .safetensors URLs from the Internet. For example, 'fofr/flux-pixar-cars'
- lora_scale
- Determines how strongly the main LoRA should be applied. Sane results between 0 and 1 for base inference. For go_fast we apply a 1.5x multiplier to this value; we've generally seen good performance when scaling the base value by that amount. You may still need to experiment to find the best value for your particular lora.
- megapixels
- Approximate number of megapixels for generated image
- num_outputs
- Number of outputs to generate
- aspect_ratio
- Aspect ratio for the generated image. If custom is selected, uses height and width below & will run in bf16 mode
- output_format
- Format of the output images
- guidance_scale
- Guidance scale for the diffusion process. Lower values can give more realistic images. Good values to try are 2, 2.5, 3 and 3.5
- output_quality
- Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs
- prompt_strength
- Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image
- extra_lora_scale
- Determines how strongly the extra LoRA should be applied. Sane results between 0 and 1 for base inference. For go_fast we apply a 1.5x multiplier to this value; we've generally seen good performance when scaling the base value by that amount. You may still need to experiment to find the best value for your particular lora.
- replicate_weights
- Load LoRA weights. Supports Replicate models in the format <owner>/<username> or <owner>/<username>/<version>, HuggingFace URLs in the format huggingface.co/<owner>/<model-name>, CivitAI URLs in the format civitai.com/models/<id>[/<model-name>], or arbitrary .safetensors URLs from the Internet. For example, 'fofr/flux-pixar-cars'
- num_inference_steps
- Number of denoising steps. More steps can give more detailed images, but take longer.
- disable_safety_checker
- Disable safety checker for generated images.
Output Schema
Output
Example Execution Logs
2025-01-10 13:46:40.275 | DEBUG | fp8.lora_loading:apply_lora_to_model:574 - Extracting keys 2025-01-10 13:46:40.275 | DEBUG | fp8.lora_loading:apply_lora_to_model:581 - Keys extracted Applying LoRA: 0%| | 0/304 [00:00<?, ?it/s] Applying LoRA: 91%|█████████ | 276/304 [00:00<00:00, 2758.09it/s] Applying LoRA: 100%|██████████| 304/304 [00:00<00:00, 2682.66it/s] 2025-01-10 13:46:40.389 | SUCCESS | fp8.lora_loading:unload_loras:564 - LoRAs unloaded in 0.11s free=29146171588608 Downloading weights 2025-01-10T13:46:40Z | INFO | [ Initiating ] chunk_size=150M dest=/tmp/tmpzbzs1r7x/weights url=https://replicate.delivery/xezq/Gk7KeWbseOkGj0HhMsDNp5VNMznVEDj9S9ZboGYX3pDYcRDUA/trained_model.tar 2025-01-10T13:46:42Z | INFO | [ Complete ] dest=/tmp/tmpzbzs1r7x/weights size="172 MB" total_elapsed=1.647s url=https://replicate.delivery/xezq/Gk7KeWbseOkGj0HhMsDNp5VNMznVEDj9S9ZboGYX3pDYcRDUA/trained_model.tar Downloaded weights in 1.67s 2025-01-10 13:46:42.063 | INFO | fp8.lora_loading:convert_lora_weights:498 - Loading LoRA weights for /src/weights-cache/5fe1387243327e8f 2025-01-10 13:46:42.139 | INFO | fp8.lora_loading:convert_lora_weights:519 - LoRA weights loaded 2025-01-10 13:46:42.139 | DEBUG | fp8.lora_loading:apply_lora_to_model:574 - Extracting keys 2025-01-10 13:46:42.139 | DEBUG | fp8.lora_loading:apply_lora_to_model:581 - Keys extracted Applying LoRA: 0%| | 0/304 [00:00<?, ?it/s] Applying LoRA: 91%|█████████ | 277/304 [00:00<00:00, 2757.86it/s] Applying LoRA: 100%|██████████| 304/304 [00:00<00:00, 2685.06it/s] 2025-01-10 13:46:42.253 | SUCCESS | fp8.lora_loading:load_lora:539 - LoRA applied in 0.19s Using seed: 51405 Image detected - setting to img2img mode Input image size: 1170x1133 Input image size set to: 1168x1136 0it [00:00, ?it/s] 1it [00:00, 6.61it/s] 2it [00:00, 4.64it/s] 3it [00:00, 4.23it/s] 4it [00:00, 4.07it/s] 5it [00:01, 3.97it/s] 6it [00:01, 3.91it/s] 7it [00:01, 3.88it/s] 8it [00:01, 3.87it/s] 9it [00:02, 3.85it/s] 10it [00:02, 3.84it/s] 11it [00:02, 3.82it/s] 12it [00:03, 3.82it/s] 13it [00:03, 3.82it/s] 14it [00:03, 3.82it/s] 15it [00:03, 3.81it/s] 16it [00:04, 3.81it/s] 17it [00:04, 3.81it/s] 18it [00:04, 3.81it/s] 19it [00:04, 3.81it/s] 20it [00:05, 3.81it/s] 21it [00:05, 3.82it/s] 22it [00:05, 3.82it/s] 23it [00:05, 3.81it/s] 23it [00:05, 3.89it/s] Total safe images: 1 out of 1
Version Details
- Version ID
9fd2577dc343ed5d78b6c6ac7aa8b6d6613f733f7db8335ad259b8d0f4adaeaa- Version Created
- January 9, 2025