jonghwan-seo/byumgyu 🖼️🔢❓📝✓ → 🖼️

▶️ 74 runs 📅 Dec 2024 ⚙️ Cog 0.11.1
lora lora-person text-to-image

About

Example Output

Prompt:

"A magical scene inspired by the Harry Potter universe featuring BYUMGYU as the central character. BYUMGYU is a young wizard wearing a majestic black robe with silver and emerald accents, holding a glowing wand emitting golden sparks. His face is looking directly at the viewer with a confident expression, his black hair slightly tousled by an unseen wind. Behind him, an ancient, grand castle with towering spires and glowing windows is visible under a starry night sky. The environment is surrounded by mystical fog, floating candles, and enchanted books hovering in the air. The image is lit by moonlight and the golden glow of his wand, creating a magical, high-contrast atmosphere."

Output

Example output

Performance Metrics

9.59s Prediction Time
9.68s Total Time
All Input Parameters
{
  "model": "dev",
  "prompt": "A magical scene inspired by the Harry Potter universe featuring BYUMGYU as the central character. BYUMGYU is a young wizard wearing a majestic black robe with silver and emerald accents, holding a glowing wand emitting golden sparks. His face is looking directly at the viewer with a confident expression, his black hair slightly tousled by an unseen wind. Behind him, an ancient, grand castle with towering spires and glowing windows is visible under a starry night sky. The environment is surrounded by mystical fog, floating candles, and enchanted books hovering in the air. The image is lit by moonlight and the golden glow of his wand, creating a magical, high-contrast atmosphere.",
  "go_fast": false,
  "lora_scale": 1,
  "megapixels": "1",
  "num_outputs": 1,
  "aspect_ratio": "16:9",
  "output_format": "webp",
  "guidance_scale": 3,
  "output_quality": 80,
  "prompt_strength": 0.8,
  "extra_lora_scale": 1,
  "num_inference_steps": 28
}
Input Parameters
mask Type: string
Image mask for image inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
seed Type: integer
Random seed. Set for reproducible generation
image Type: string
Input image for image to image or inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
model Default: dev
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 Type: integerRange: 256 - 1440
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 Type: integerRange: 256 - 1440
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) Type: string
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 Type: booleanDefault: false
Run faster predictions with model optimized for speed (currently fp8 quantized); disable to run in original bf16
extra_lora Type: string
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 Type: numberDefault: 1Range: -1 - 3
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 Default: 1
Approximate number of megapixels for generated image
num_outputs Type: integerDefault: 1Range: 1 - 4
Number of outputs to generate
aspect_ratio Default: 1:1
Aspect ratio for the generated image. If custom is selected, uses height and width below & will run in bf16 mode
output_format Default: webp
Format of the output images
guidance_scale Type: numberDefault: 3Range: 0 - 10
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 Type: integerDefault: 80Range: 0 - 100
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 Type: numberDefault: 0.8Range: 0 - 1
Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image
extra_lora_scale Type: numberDefault: 1Range: -1 - 3
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 Type: string
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 Type: integerDefault: 28Range: 1 - 50
Number of denoising steps. More steps can give more detailed images, but take longer.
disable_safety_checker Type: booleanDefault: false
Disable safety checker for generated images.
Output Schema

Output

Type: arrayItems Type: stringItems Format: uri

Example Execution Logs
2024-12-06 16:08:11.373 | DEBUG    | fp8.lora_loading:apply_lora_to_model:569 - Extracting keys
2024-12-06 16:08:11.373 | DEBUG    | fp8.lora_loading:apply_lora_to_model:576 - Keys extracted
Applying LoRA:   0%|          | 0/304 [00:00<?, ?it/s]
Applying LoRA:  97%|█████████▋| 295/304 [00:00<00:00, 2943.92it/s]
Applying LoRA: 100%|██████████| 304/304 [00:00<00:00, 2854.14it/s]
2024-12-06 16:08:11.480 | SUCCESS  | fp8.lora_loading:unload_loras:559 - LoRAs unloaded in 0.11s
free=29022338072576
Downloading weights
2024-12-06T16:08:11Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmp8_75998e/weights url=https://replicate.delivery/xezq/kKNQQCOfah2ye01OLqIkDiSewX2683j2a7fCyHfbrWmI2jBfE/trained_model.tar
2024-12-06T16:08:14Z | INFO  | [ Complete ] dest=/tmp/tmp8_75998e/weights size="172 MB" total_elapsed=3.164s url=https://replicate.delivery/xezq/kKNQQCOfah2ye01OLqIkDiSewX2683j2a7fCyHfbrWmI2jBfE/trained_model.tar
Downloaded weights in 3.19s
2024-12-06 16:08:14.670 | INFO     | fp8.lora_loading:convert_lora_weights:493 - Loading LoRA weights for /src/weights-cache/fc493b5fc13a19fa
2024-12-06 16:08:14.740 | INFO     | fp8.lora_loading:convert_lora_weights:514 - LoRA weights loaded
2024-12-06 16:08:14.740 | DEBUG    | fp8.lora_loading:apply_lora_to_model:569 - Extracting keys
2024-12-06 16:08:14.740 | DEBUG    | fp8.lora_loading:apply_lora_to_model:576 - Keys extracted
Applying LoRA:   0%|          | 0/304 [00:00<?, ?it/s]
Applying LoRA:  97%|█████████▋| 295/304 [00:00<00:00, 2949.84it/s]
Applying LoRA: 100%|██████████| 304/304 [00:00<00:00, 2859.75it/s]
2024-12-06 16:08:14.847 | SUCCESS  | fp8.lora_loading:load_lora:534 - LoRA applied in 0.18s
Using seed: 14598
0it [00:00, ?it/s]
1it [00:00,  8.45it/s]
2it [00:00,  5.89it/s]
3it [00:00,  5.36it/s]
4it [00:00,  5.14it/s]
5it [00:00,  5.01it/s]
6it [00:01,  4.92it/s]
7it [00:01,  4.89it/s]
8it [00:01,  4.87it/s]
9it [00:01,  4.86it/s]
10it [00:01,  4.84it/s]
11it [00:02,  4.82it/s]
12it [00:02,  4.81it/s]
13it [00:02,  4.81it/s]
14it [00:02,  4.81it/s]
15it [00:03,  4.81it/s]
16it [00:03,  4.80it/s]
17it [00:03,  4.80it/s]
18it [00:03,  4.81it/s]
19it [00:03,  4.81it/s]
20it [00:04,  4.80it/s]
21it [00:04,  4.80it/s]
22it [00:04,  4.80it/s]
23it [00:04,  4.81it/s]
24it [00:04,  4.80it/s]
25it [00:05,  4.80it/s]
26it [00:05,  4.79it/s]
27it [00:05,  4.79it/s]
28it [00:05,  4.80it/s]
28it [00:05,  4.88it/s]
Total safe images: 1 out of 1
Version Details
Version ID
b37b7fcd5a4bb42d840fe68b525e2a991c2a9df288269fde8c9cf5235a59a2ca
Version Created
December 6, 2024
Run on Replicate →