animacustoms/ernestamod 🖼️🔢❓📝✓ → 🖼️

▶️ 77 runs 📅 Nov 2024 ⚙️ Cog 0.11.1
image-inpainting image-to-image lora text-to-image

About

Example Output

Prompt:

"(stunning 18 years old woman), confidently sitting in luxurious Bentley Continental GT, dressed in elegant (sleek black dress with thigh-high slit and open back), detailed facial features and expression, moody evening lighting, high-definition, warm and inviting ambiance, showcasing opulence and style, surrounding luxurious interior details, ultra-detailed, capturing confidence and youth in a sophisticated atmosphere."

Output

Example outputExample outputExample outputExample output

Performance Metrics

27.89s Prediction Time
28.20s Total Time
All Input Parameters 
{
  "model": "dev",
  "width": 1440,
  "height": 1440,
  "prompt": "(stunning 18 years old woman), confidently sitting in luxurious Bentley Continental GT, dressed in elegant (sleek black dress with thigh-high slit and open back), detailed facial features and expression, moody evening lighting, high-definition, warm and inviting ambiance, showcasing opulence and style, surrounding luxurious interior details, ultra-detailed, capturing confidence and youth in a sophisticated atmosphere.",
  "go_fast": false,
  "lora_scale": 0.94,
  "megapixels": "1",
  "num_outputs": 4,
  "aspect_ratio": "3:2",
  "output_format": "webp",
  "guidance_scale": 2.06,
  "output_quality": 100,
  "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: array • Items Type: string • Items Format: uri

Example Execution Logs 
2024-11-30 03:07:36.232 | DEBUG    | fp8.lora_loading:apply_lora_to_model:569 - Extracting keys
2024-11-30 03:07:36.233 | DEBUG    | fp8.lora_loading:apply_lora_to_model:576 - Keys extracted
Applying LoRA:   0%|          | 0/304 [00:00<?, ?it/s]
Applying LoRA:  92%|█████████▏| 279/304 [00:00<00:00, 2766.12it/s]
Applying LoRA: 100%|██████████| 304/304 [00:00<00:00, 2680.84it/s]
2024-11-30 03:07:36.346 | SUCCESS  | fp8.lora_loading:unload_loras:559 - LoRAs unloaded in 0.11s
free=29028307935232
Downloading weights
2024-11-30T03:07:36Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmpw04xloi1/weights url=https://replicate.delivery/xezq/NrbQj3eO212WMq2frewfgxa0MiADDziOFPk4E3f3lcqhQQweE/trained_model.tar
2024-11-30T03:07:37Z | INFO  | [ Complete ] dest=/tmp/tmpw04xloi1/weights size="172 MB" total_elapsed=1.625s url=https://replicate.delivery/xezq/NrbQj3eO212WMq2frewfgxa0MiADDziOFPk4E3f3lcqhQQweE/trained_model.tar
Downloaded weights in 1.65s
2024-11-30 03:07:37.996 | INFO     | fp8.lora_loading:convert_lora_weights:493 - Loading LoRA weights for /src/weights-cache/d06f933a2f66b9da
2024-11-30 03:07:38.067 | INFO     | fp8.lora_loading:convert_lora_weights:514 - LoRA weights loaded
2024-11-30 03:07:38.067 | DEBUG    | fp8.lora_loading:apply_lora_to_model:569 - Extracting keys
2024-11-30 03:07:38.067 | DEBUG    | fp8.lora_loading:apply_lora_to_model:576 - Keys extracted
Applying LoRA:   0%|          | 0/304 [00:00<?, ?it/s]
Applying LoRA:  92%|█████████▏| 279/304 [00:00<00:00, 2772.31it/s]
Applying LoRA: 100%|██████████| 304/304 [00:00<00:00, 2686.80it/s]
2024-11-30 03:07:38.181 | SUCCESS  | fp8.lora_loading:load_lora:534 - LoRA applied in 0.18s
Using seed: 55771
0it [00:00, ?it/s]
1it [00:00,  8.78it/s]
2it [00:00,  6.10it/s]
3it [00:00,  5.57it/s]
4it [00:00,  5.36it/s]
5it [00:00,  5.19it/s]
6it [00:01,  5.11it/s]
7it [00:01,  5.08it/s]
8it [00:01,  5.06it/s]
9it [00:01,  5.06it/s]
10it [00:01,  5.04it/s]
11it [00:02,  5.02it/s]
12it [00:02,  5.02it/s]
13it [00:02,  5.02it/s]
14it [00:02,  5.02it/s]
15it [00:02,  5.02it/s]
16it [00:03,  5.01it/s]
17it [00:03,  5.01it/s]
18it [00:03,  5.00it/s]
19it [00:03,  5.00it/s]
20it [00:03,  5.00it/s]
21it [00:04,  5.01it/s]
22it [00:04,  5.01it/s]
23it [00:04,  5.02it/s]
24it [00:04,  5.02it/s]
25it [00:04,  5.02it/s]
26it [00:05,  5.01it/s]
27it [00:05,  5.00it/s]
28it [00:05,  5.00it/s]
28it [00:05,  5.09it/s]
0it [00:00, ?it/s]
1it [00:00,  5.05it/s]
2it [00:00,  5.01it/s]
3it [00:00,  5.01it/s]
4it [00:00,  5.00it/s]
5it [00:00,  5.00it/s]
6it [00:01,  5.01it/s]
7it [00:01,  5.01it/s]
8it [00:01,  5.01it/s]
9it [00:01,  5.00it/s]
10it [00:01,  5.01it/s]
11it [00:02,  5.01it/s]
12it [00:02,  5.02it/s]
13it [00:02,  5.01it/s]
14it [00:02,  5.01it/s]
15it [00:02,  5.00it/s]
16it [00:03,  5.00it/s]
17it [00:03,  5.01it/s]
18it [00:03,  5.00it/s]
19it [00:03,  5.02it/s]
20it [00:03,  5.02it/s]
21it [00:04,  5.01it/s]
22it [00:04,  5.01it/s]
23it [00:04,  5.01it/s]
24it [00:04,  5.01it/s]
25it [00:04,  5.00it/s]
26it [00:05,  5.00it/s]
27it [00:05,  4.99it/s]
28it [00:05,  5.00it/s]
28it [00:05,  5.01it/s]
0it [00:00, ?it/s]
1it [00:00,  5.06it/s]
2it [00:00,  5.04it/s]
3it [00:00,  5.02it/s]
4it [00:00,  5.01it/s]
5it [00:00,  5.00it/s]
6it [00:01,  5.00it/s]
7it [00:01,  5.01it/s]
8it [00:01,  5.01it/s]
9it [00:01,  5.01it/s]
10it [00:01,  5.00it/s]
11it [00:02,  5.00it/s]
12it [00:02,  4.99it/s]
13it [00:02,  4.99it/s]
14it [00:02,  5.01it/s]
15it [00:02,  5.01it/s]
16it [00:03,  5.01it/s]
17it [00:03,  5.00it/s]
18it [00:03,  5.01it/s]
19it [00:03,  5.01it/s]
20it [00:03,  5.01it/s]
21it [00:04,  5.00it/s]
22it [00:04,  5.00it/s]
23it [00:04,  5.00it/s]
24it [00:04,  5.00it/s]
25it [00:04,  4.99it/s]
26it [00:05,  4.99it/s]
27it [00:05,  4.99it/s]
28it [00:05,  4.99it/s]
28it [00:05,  5.00it/s]
0it [00:00, ?it/s]
1it [00:00,  5.04it/s]
2it [00:00,  5.01it/s]
3it [00:00,  5.01it/s]
4it [00:00,  4.99it/s]
5it [00:01,  4.99it/s]
6it [00:01,  4.97it/s]
7it [00:01,  4.97it/s]
8it [00:01,  4.99it/s]
9it [00:01,  4.99it/s]
10it [00:02,  4.99it/s]
11it [00:02,  4.99it/s]
12it [00:02,  4.98it/s]
13it [00:02,  4.97it/s]
14it [00:02,  4.98it/s]
15it [00:03,  4.99it/s]
16it [00:03,  4.99it/s]
17it [00:03,  4.99it/s]
18it [00:03,  4.99it/s]
19it [00:03,  4.99it/s]
20it [00:04,  4.99it/s]
21it [00:04,  4.99it/s]
22it [00:04,  4.99it/s]
23it [00:04,  4.99it/s]
24it [00:04,  4.99it/s]
25it [00:05,  4.99it/s]
26it [00:05,  4.99it/s]
27it [00:05,  4.99it/s]
28it [00:05,  4.99it/s]
28it [00:05,  4.99it/s]
Total safe images: 4 out of 4
Version Details
Version ID
e811b8022bf1886d6edef817bef9b40e876c79377eec652db351431fc1859901
Version Created
November 30, 2024
Run on Replicate →