zzyjay/icedevsfm 🖼️🔢❓📝✓ → 🖼️

▶️ 164 runs 📅 Sep 2024 ⚙️ Cog 0.9.20

About

Example Output

Prompt:

"realism:in studio,white background:an icedev male monster,gaint,tall and strong,open mouth widely,Bull Head-male"

Output

Performance Metrics

16.56s Prediction Time

24.14s Total Time

All Input Parameters

{
  "model": "dev",
  "prompt": "realism:in studio,white background:an icedev male monster,gaint,tall and strong,open mouth widely,Bull Head-male",
  "go_fast": false,
  "lora_scale": 1.09,
  "megapixels": "1",
  "num_outputs": 1,
  "aspect_ratio": "9:21",
  "output_format": "jpg",
  "guidance_scale": 3,
  "output_quality": 100,
  "prompt_strength": 0.8,
  "extra_lora_scale": 1,
  "num_inference_steps": 41
}

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

free=26013647654912
Downloading weights
2025-04-23T13:57:16Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmppovjxydt/weights url=https://replicate.delivery/yhqm/uiNW5AQoL0aqOR4ensvKi4kmFMadz0R6exTppjRD0utUJjZTA/trained_model.tar
2025-04-23T13:57:16Z | INFO  | [ Cache Service ] enabled=true scheme=http target=hermes.services.svc.cluster.local
2025-04-23T13:57:22Z | INFO  | [ Complete ] dest=/tmp/tmppovjxydt/weights size="172 MB" total_elapsed=5.600s url=https://replicate.delivery/yhqm/uiNW5AQoL0aqOR4ensvKi4kmFMadz0R6exTppjRD0utUJjZTA/trained_model.tar
Downloaded weights in 5.62s
Loaded LoRAs in 6.19s
Using seed: 35607
Prompt: realism:in studio,white background:an icedev male monster,gaint,tall and strong,open mouth widely,Bull Head-male
[!] txt2img mode
  0%|          | 0/41 [00:00<?, ?it/s]
  2%|▏         | 1/41 [00:00<00:10,  3.94it/s]
  5%|▍         | 2/41 [00:00<00:08,  4.35it/s]
  7%|▋         | 3/41 [00:00<00:09,  4.20it/s]
 10%|▉         | 4/41 [00:00<00:08,  4.12it/s]
 12%|█▏        | 5/41 [00:01<00:08,  4.09it/s]
 15%|█▍        | 6/41 [00:01<00:08,  4.07it/s]
 17%|█▋        | 7/41 [00:01<00:08,  4.05it/s]
 20%|█▉        | 8/41 [00:01<00:08,  4.05it/s]
 22%|██▏       | 9/41 [00:02<00:07,  4.04it/s]
 24%|██▍       | 10/41 [00:02<00:07,  4.03it/s]
 27%|██▋       | 11/41 [00:02<00:07,  4.03it/s]
 29%|██▉       | 12/41 [00:02<00:07,  4.03it/s]
 32%|███▏      | 13/41 [00:03<00:06,  4.03it/s]
 34%|███▍      | 14/41 [00:03<00:06,  4.03it/s]
 37%|███▋      | 15/41 [00:03<00:06,  4.02it/s]
 39%|███▉      | 16/41 [00:03<00:06,  4.02it/s]
 41%|████▏     | 17/41 [00:04<00:05,  4.02it/s]
 44%|████▍     | 18/41 [00:04<00:05,  4.02it/s]
 46%|████▋     | 19/41 [00:04<00:05,  4.02it/s]
 49%|████▉     | 20/41 [00:04<00:05,  4.02it/s]
 51%|█████     | 21/41 [00:05<00:04,  4.03it/s]
 54%|█████▎    | 22/41 [00:05<00:04,  4.03it/s]
 56%|█████▌    | 23/41 [00:05<00:04,  4.02it/s]
 59%|█████▊    | 24/41 [00:05<00:04,  4.02it/s]
 61%|██████    | 25/41 [00:06<00:03,  4.02it/s]
 63%|██████▎   | 26/41 [00:06<00:03,  4.02it/s]
 66%|██████▌   | 27/41 [00:06<00:03,  4.02it/s]
 68%|██████▊   | 28/41 [00:06<00:03,  4.02it/s]
 71%|███████   | 29/41 [00:07<00:02,  4.02it/s]
 73%|███████▎  | 30/41 [00:07<00:02,  4.02it/s]
 76%|███████▌  | 31/41 [00:07<00:02,  4.03it/s]
 78%|███████▊  | 32/41 [00:07<00:02,  4.02it/s]
 80%|████████  | 33/41 [00:08<00:01,  4.03it/s]
 83%|████████▎ | 34/41 [00:08<00:01,  4.03it/s]
 85%|████████▌ | 35/41 [00:08<00:01,  4.02it/s]
 88%|████████▊ | 36/41 [00:08<00:01,  4.02it/s]
 90%|█████████ | 37/41 [00:09<00:00,  4.02it/s]
 93%|█████████▎| 38/41 [00:09<00:00,  4.02it/s]
 95%|█████████▌| 39/41 [00:09<00:00,  4.03it/s]
 98%|█████████▊| 40/41 [00:09<00:00,  4.02it/s]
100%|██████████| 41/41 [00:10<00:00,  4.03it/s]
100%|██████████| 41/41 [00:10<00:00,  4.04it/s]
Total safe images: 1 out of 1

Version Details

Version ID: e57c2dfbc48a476779abad3b6695839ecb779c18d0ec95f16d1f677a99cb3a42
Version Created: September 4, 2024

Run on Replicate →