biguloff/alanskoe-male-character 🖼️🔢❓📝✓ → 🖼️

▶️ 35 runs 📅 Nov 2024 ⚙️ Cog 0.11.1
image-inpainting image-to-image lora lora-person text-to-image

About

Example Output

Prompt:

"

  1. Subject:
    A dimly lit stone hall with towering vaulted ceilings. In the center of the hall, King Uruzmag TOK sits on a massive, intricately carved stone throne, surrounded by elders arranged in a semi-circle. King Uruzmag is a tall man of 45-50 years, with sharp facial features, a thick graying beard, and piercing brown eyes that convey focus and inner strength. He is dressed in a black burka with golden embroidery, a vivid red sash holding a ceremonial dagger at his side, and a small crown adorned with Ossetian symbols rests on his head. His posture exudes authority, while his expression reflects concentration and subtle tension.

  2. Style:
    Cinematic realism with a dramatic tone, inspired by "The Lord of the Rings" and "Dune." The scene emphasizes the grandeur and tension of the moment, with a focus on historical accuracy and cultural richness.

  3. Setting:
    A royal chamber in 980s Ossetia. The hall is constructed from dark, weathered stone, with high, arched ceilings that fade into shadow. The room is dimly lit by torches mounted on the walls, their flickering flames casting dynamic, sharp shadows across the cracked stone surfaces. The throne platform is slightly elevated, drawing attention to the king as the central figure.

  4. Composition:
    Wide shot with a static frame to capture the full scale of the scene. The foreground includes the cracked stone floor leading to the midground, where King Uruzmag is seated on his throne. The semi-circle of elders is positioned around the king, their postures and expressions faintly visible in the firelight. The background includes towering stone walls lined with torches, the flames providing a warm contrast to the cold stone environment.

  5. Lighting:Motion blur
    Warm, flickering torchlight highlights the throne and King Uruzmag, casting intricate shadows that emphasize the texture of his cloak, the golden embroidery, and the details of his crown. The elders are partially illuminated, with soft shadows adding depth to their forms. The corners of the hall are shrouded in darkness, creating a dramatic, tense atmosphere.

  6. Additional Details:
    The cracked stone floor has faint engravings and natural imperfections, adding texture and realism. The throne is carved with traditional Ossetian symbols, exuding an air of authority and heritage. The flickering light plays across the king’s face, accentuating his focused expression and graying beard. The elders wear traditional garments in muted tones, their faces lined with age and concern. The scene is imbued with solemnity, with the stillness of the characters conveying the weight of the moment.

"

Output

Example outputExample output

Performance Metrics

12.35s Prediction Time
12.40s Total Time
All Input Parameters 
{
  "model": "dev",
  "prompt": "1. Subject: \nA dimly lit stone hall with towering vaulted ceilings. In the center of the hall, King Uruzmag TOK sits on a massive, intricately carved stone throne, surrounded by elders arranged in a semi-circle. King Uruzmag is a tall man of 45-50 years, with sharp facial features, a thick graying beard, and piercing brown eyes that convey focus and inner strength. He is dressed in a black burka with golden embroidery, a vivid red sash holding a ceremonial dagger at his side, and a small crown adorned with Ossetian symbols rests on his head. His posture exudes authority, while his expression reflects concentration and subtle tension.\n\n2. Style:\nCinematic realism with a dramatic tone, inspired by \"The Lord of the Rings\" and \"Dune.\" The scene emphasizes the grandeur and tension of the moment, with a focus on historical accuracy and cultural richness.\n\n3. Setting:\nA royal chamber in 980s Ossetia. The hall is constructed from dark, weathered stone, with high, arched ceilings that fade into shadow. The room is dimly lit by torches mounted on the walls, their flickering flames casting dynamic, sharp shadows across the cracked stone surfaces. The throne platform is slightly elevated, drawing attention to the king as the central figure.\n\n4. Composition:\nWide shot with a static frame to capture the full scale of the scene. The foreground includes the cracked stone floor leading to the midground, where King Uruzmag is seated on his throne. The semi-circle of elders is positioned around the king, their postures and expressions faintly visible in the firelight. The background includes towering stone walls lined with torches, the flames providing a warm contrast to the cold stone environment.\n\n5. Lighting:Motion blur\nWarm, flickering torchlight highlights the throne and King Uruzmag, casting intricate shadows that emphasize the texture of his cloak, the golden embroidery, and the details of his crown. The elders are partially illuminated, with soft shadows adding depth to their forms. The corners of the hall are shrouded in darkness, creating a dramatic, tense atmosphere.\n\n6. Additional Details:\nThe cracked stone floor has faint engravings and natural imperfections, adding texture and realism. The throne is carved with traditional Ossetian symbols, exuding an air of authority and heritage. The flickering light plays across the king’s face, accentuating his focused expression and graying beard. The elders wear traditional garments in muted tones, their faces lined with age and concern. The scene is imbued with solemnity, with the stillness of the characters conveying the weight of the moment.",
  "go_fast": true,
  "extra_lora": "https://civitai.com/api/download/models/794602?type=Model&format=SafeTensor&token=3897b44086a90075dbac7bf6ba7a94ec",
  "lora_scale": 0.53,
  "megapixels": "1",
  "num_outputs": 2,
  "aspect_ratio": "16:9",
  "output_format": "png",
  "guidance_scale": 3,
  "output_quality": 80,
  "prompt_strength": 0.8,
  "extra_lora_scale": 0.52,
  "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 
2024-11-27 09:33:02.714 | DEBUG    | fp8.lora_loading:apply_lora_to_model:569 - Extracting keys
2024-11-27 09:33:02.714 | DEBUG    | fp8.lora_loading:apply_lora_to_model:576 - Keys extracted
Applying LoRA:   0%|          | 0/304 [00:00<?, ?it/s]
Applying LoRA: 100%|██████████| 304/304 [00:00<00:00, 12498.47it/s]
2024-11-27 09:33:02.739 | SUCCESS  | fp8.lora_loading:unload_loras:559 - LoRAs unloaded in 0.025s
free=29255247814656
Downloading weights
2024-11-27T09:33:02Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmpkce3ctrt/weights url=https://replicate.delivery/xezq/BpSnxiJlwqoiD96jGGePfFznaOkLAWVagurJEFd7L8ze9JmnA/trained_model.tar
2024-11-27T09:33:05Z | INFO  | [ Complete ] dest=/tmp/tmpkce3ctrt/weights size="344 MB" total_elapsed=2.311s url=https://replicate.delivery/xezq/BpSnxiJlwqoiD96jGGePfFznaOkLAWVagurJEFd7L8ze9JmnA/trained_model.tar
Downloaded weights in 2.36s
free=29254901698560
Downloading weights
2024-11-27T09:33:05Z | INFO  | [ Initiating ] chunk_size=150M dest=/src/weights-cache/49e84ada1b7b739f url=https://civitai.com/api/download/models/794602?type=Model&format=SafeTensor&token=3897b44086a90075dbac7bf6ba7a94ec
2024-11-27T09:33:05Z | INFO  | [ Redirect ] redirect_url=https://civitai-delivery-worker-prod.5ac0637cfd0766c97916cefa3764fbdf.r2.cloudflarestorage.com/model/2461572/fluxDuneCinematic.Ma1I.safetensors?X-Amz-Expires=86400&response-content-disposition=attachment%3B%20filename%3D%22Flux_Dune_Cinematic.safetensors%22&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=e01358d793ad6966166af8b3064953ad/20241127/us-east-1/s3/aws4_request&X-Amz-Date=20241127T093305Z&X-Amz-SignedHeaders=host&X-Amz-Signature=4a89a500dd593d71bdee5d2c9454d85a703b57c75336c8b71512427f63c755c8 url=https://civitai.com/api/download/models/794602?type=Model&format=SafeTensor&token=3897b44086a90075dbac7bf6ba7a94ec
2024-11-27T09:33:07Z | INFO  | [ Complete ] dest=/src/weights-cache/49e84ada1b7b739f size="172 MB" total_elapsed=2.635s url=https://civitai.com/api/download/models/794602?type=Model&format=SafeTensor&token=3897b44086a90075dbac7bf6ba7a94ec
Downloaded weights in 2.66s
2024-11-27 09:33:07.757 | INFO     | fp8.lora_loading:convert_lora_weights:493 - Loading LoRA weights for /src/weights-cache/4e0cec0fccc93d15
2024-11-27 09:33:07.858 | INFO     | fp8.lora_loading:convert_lora_weights:514 - LoRA weights loaded
2024-11-27 09:33:07.858 | DEBUG    | fp8.lora_loading:apply_lora_to_model:569 - Extracting keys
2024-11-27 09:33:07.858 | DEBUG    | fp8.lora_loading:apply_lora_to_model:576 - Keys extracted
Applying LoRA:   0%|          | 0/304 [00:00<?, ?it/s]
Applying LoRA:  41%|████      | 125/304 [00:00<00:00, 1230.81it/s]
Applying LoRA:  82%|████████▏ | 249/304 [00:00<00:00, 933.43it/s] 
Applying LoRA: 100%|██████████| 304/304 [00:00<00:00, 956.75it/s]
2024-11-27 09:33:08.176 | SUCCESS  | fp8.lora_loading:load_lora:534 - LoRA applied in 0.42s
2024-11-27 09:33:08.176 | INFO     | fp8.lora_loading:convert_lora_weights:493 - Loading LoRA weights for /src/weights-cache/49e84ada1b7b739f
2024-11-27 09:33:08.285 | INFO     | fp8.lora_loading:convert_lora_weights:514 - LoRA weights loaded
2024-11-27 09:33:08.286 | DEBUG    | fp8.lora_loading:apply_lora_to_model:569 - Extracting keys
2024-11-27 09:33:08.286 | DEBUG    | fp8.lora_loading:apply_lora_to_model:576 - Keys extracted
Applying LoRA:   0%|          | 0/304 [00:00<?, ?it/s]
Applying LoRA:  41%|████      | 125/304 [00:00<00:00, 1231.95it/s]
Applying LoRA:  82%|████████▏ | 249/304 [00:00<00:00, 934.23it/s] 
Applying LoRA: 100%|██████████| 304/304 [00:00<00:00, 957.47it/s]
2024-11-27 09:33:08.604 | SUCCESS  | fp8.lora_loading:load_lora:534 - LoRA applied in 0.43s
running quantized prediction
Using seed: 142681503
  0%|          | 0/28 [00:00<?, ?it/s]
  7%|▋         | 2/28 [00:00<00:01, 15.78it/s]
 14%|█▍        | 4/28 [00:00<00:02, 11.91it/s]
 21%|██▏       | 6/28 [00:00<00:01, 11.11it/s]
 29%|██▊       | 8/28 [00:00<00:01, 10.68it/s]
 36%|███▌      | 10/28 [00:00<00:01, 10.43it/s]
 43%|████▎     | 12/28 [00:01<00:01, 10.27it/s]
 50%|█████     | 14/28 [00:01<00:01, 10.24it/s]
 57%|█████▋    | 16/28 [00:01<00:01, 10.26it/s]
 64%|██████▍   | 18/28 [00:01<00:00, 10.21it/s]
 71%|███████▏  | 20/28 [00:01<00:00, 10.15it/s]
 79%|███████▊  | 22/28 [00:02<00:00, 10.10it/s]
 86%|████████▌ | 24/28 [00:02<00:00, 10.07it/s]
 93%|█████████▎| 26/28 [00:02<00:00, 10.07it/s]
100%|██████████| 28/28 [00:02<00:00, 10.12it/s]
100%|██████████| 28/28 [00:02<00:00, 10.37it/s]
  0%|          | 0/28 [00:00<?, ?it/s]
  7%|▋         | 2/28 [00:00<00:02, 10.04it/s]
 14%|█▍        | 4/28 [00:00<00:02, 10.04it/s]
 21%|██▏       | 6/28 [00:00<00:02, 10.02it/s]
 29%|██▊       | 8/28 [00:00<00:01, 10.02it/s]
 36%|███▌      | 10/28 [00:00<00:01, 10.06it/s]
 43%|████▎     | 12/28 [00:01<00:01, 10.09it/s]
 50%|█████     | 14/28 [00:01<00:01, 10.09it/s]
 57%|█████▋    | 16/28 [00:01<00:01, 10.10it/s]
 64%|██████▍   | 18/28 [00:01<00:00, 10.09it/s]
 71%|███████▏  | 20/28 [00:01<00:00, 10.07it/s]
 79%|███████▊  | 22/28 [00:02<00:00, 10.06it/s]
 86%|████████▌ | 24/28 [00:02<00:00, 10.05it/s]
 93%|█████████▎| 26/28 [00:02<00:00, 10.05it/s]
100%|██████████| 28/28 [00:02<00:00, 10.10it/s]
100%|██████████| 28/28 [00:02<00:00, 10.07it/s]
Total safe images: 2 out of 2
Version Details
Version ID
0ba4c35f90c68566bebd5f7fd27563dd58b3f7b2b336869474e2789327152e07
Version Created
November 21, 2024
Run on Replicate →