patternstech/raas-replicate 🖼️🔢❓📝✓ → 🖼️

▶️ 8 runs 📅 Nov 2025 ⚙️ Cog 0.15.3

About

Example Output

Prompt:

"A hyper-realistic image
of raas in a dark podcast studio. He is wearing a black
sweatshirt. The lighting is low and
moody, with neon accents softly illuminating the scene. The man
is looking directly at the camera with a neutral expression. His
head is proportionate and normal-sized. The camera is zoomed
way out, capturing a wide shot that includes his full upper body,
the desk, microphone setup, and surrounding negative space. The
background is minimal—no posters, shelves, or decorations on
the walls—just clean darkness with subtle neon glow. The style is
cinematic and photorealistic. Head & toros are both proportional.
toros looks athletic and slim. skin is fair. make it front pose. face also has a jawline. no dark circles. fair skin tone. hirs combed back. "

Output

Performance Metrics

22.51s Prediction Time

22.60s Total Time

All Input Parameters

{
  "model": "dev",
  "prompt": "A hyper-realistic image\nof raas in a dark podcast studio. He is wearing a black\nsweatshirt. The lighting is low and\nmoody, with neon accents softly illuminating the scene. The man\nis looking directly at the camera with a neutral expression. His\nhead is proportionate and normal-sized. The camera is zoomed\nway out, capturing a wide shot that includes his full upper body,\nthe desk, microphone setup, and surrounding negative space. The\nbackground is minimal—no posters, shelves, or decorations on\nthe walls—just clean darkness with subtle neon glow. The style is\ncinematic and photorealistic. Head & toros are both proportional.\ntoros looks athletic and slim. skin is fair. make it front pose. face also has a jawline. no dark circles. fair skin tone. hirs combed back. \n",
  "go_fast": false,
  "lora_scale": 1,
  "megapixels": "1",
  "num_outputs": 2,
  "aspect_ratio": "1:1",
  "output_format": "jpg",
  "guidance_scale": 3.64,
  "output_quality": 80,
  "prompt_strength": 0.8,
  "extra_lora_scale": 1,
  "num_inference_steps": 34
}

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=23626669330432
Downloading weights
2025-11-05T06:27:33Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmp17v4mmtw/weights url=https://replicate.delivery/xezq/MYwt5aFpfD2CT6GMGAdUEP6XfpPQ8NEkddjRSRSK5opQrJmVA/trained_model.tar
2025-11-05T06:27:33Z | INFO  | [ Cache Service ] enabled=true scheme=http target=hermes.services.svc.cluster.local
2025-11-05T06:27:33Z | INFO  | [ Cache URL Rewrite ] enabled=true target_url=http://hermes.services.svc.cluster.local/replicate.delivery/xezq/MYwt5aFpfD2CT6GMGAdUEP6XfpPQ8NEkddjRSRSK5opQrJmVA/trained_model.tar url=https://replicate.delivery/xezq/MYwt5aFpfD2CT6GMGAdUEP6XfpPQ8NEkddjRSRSK5opQrJmVA/trained_model.tar
2025-11-05T06:27:33Z | INFO  | [ Redirect ] redirect_url=http://r8-east4-loras-ric1.cwlota.com/694788dc684c586ff82f2f41315bbb2c9c9ca2246a76251b68497abc81ca60d6?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Checksum-Mode=ENABLED&X-Amz-Credential=CWNZUVKLDHXVHEZN%2F20251105%2FUS-EAST-04A%2Fs3%2Faws4_request&X-Amz-Date=20251105T062733Z&X-Amz-Expires=600&X-Amz-SignedHeaders=host&x-id=GetObject&X-Amz-Signature=475fb285c5ef3dec8ddf8f86d23f02307a4bd185db3a709910640a402c7a2a11 url=http://hermes.services.svc.cluster.local/replicate.delivery/xezq/MYwt5aFpfD2CT6GMGAdUEP6XfpPQ8NEkddjRSRSK5opQrJmVA/trained_model.tar
2025-11-05T06:27:33Z | INFO  | [ Complete ] dest=/tmp/tmp17v4mmtw/weights size="483 MB" total_elapsed=0.501s url=https://replicate.delivery/xezq/MYwt5aFpfD2CT6GMGAdUEP6XfpPQ8NEkddjRSRSK5opQrJmVA/trained_model.tar
Downloaded weights in 0.59s
Loaded LoRAs in 1.59s
Using seed: 18059
Prompt: A hyper-realistic image
of raas in a dark podcast studio. He is wearing a black
sweatshirt. The lighting is low and
moody, with neon accents softly illuminating the scene. The man
is looking directly at the camera with a neutral expression. His
head is proportionate and normal-sized. The camera is zoomed
way out, capturing a wide shot that includes his full upper body,
the desk, microphone setup, and surrounding negative space. The
background is minimal—no posters, shelves, or decorations on
the walls—just clean darkness with subtle neon glow. The style is
cinematic and photorealistic. Head & toros are both proportional.
toros looks athletic and slim. skin is fair. make it front pose. face also has a jawline. no dark circles. fair skin tone. hirs combed back.
[!] txt2img mode
  0%|          | 0/34 [00:00<?, ?it/s]
  3%|▎         | 1/34 [00:00<00:18,  1.80it/s]
  6%|▌         | 2/34 [00:00<00:15,  2.06it/s]
  9%|▉         | 3/34 [00:01<00:15,  1.94it/s]
 12%|█▏        | 4/34 [00:02<00:15,  1.89it/s]
 15%|█▍        | 5/34 [00:02<00:15,  1.86it/s]
 18%|█▊        | 6/34 [00:03<00:15,  1.85it/s]
 21%|██        | 7/34 [00:03<00:14,  1.84it/s]
 24%|██▎       | 8/34 [00:04<00:14,  1.83it/s]
 26%|██▋       | 9/34 [00:04<00:13,  1.82it/s]
 29%|██▉       | 10/34 [00:05<00:13,  1.82it/s]
 32%|███▏      | 11/34 [00:05<00:12,  1.82it/s]
 35%|███▌      | 12/34 [00:06<00:12,  1.82it/s]
 38%|███▊      | 13/34 [00:07<00:11,  1.81it/s]
 41%|████      | 14/34 [00:07<00:11,  1.81it/s]
 44%|████▍     | 15/34 [00:08<00:10,  1.81it/s]
 47%|████▋     | 16/34 [00:08<00:09,  1.81it/s]
 50%|█████     | 17/34 [00:09<00:09,  1.81it/s]
 53%|█████▎    | 18/34 [00:09<00:08,  1.81it/s]
 56%|█████▌    | 19/34 [00:10<00:08,  1.81it/s]
 59%|█████▉    | 20/34 [00:10<00:07,  1.81it/s]
 62%|██████▏   | 21/34 [00:11<00:07,  1.81it/s]
 65%|██████▍   | 22/34 [00:12<00:06,  1.81it/s]
 68%|██████▊   | 23/34 [00:12<00:06,  1.81it/s]
 71%|███████   | 24/34 [00:13<00:05,  1.81it/s]
 74%|███████▎  | 25/34 [00:13<00:04,  1.81it/s]
 76%|███████▋  | 26/34 [00:14<00:04,  1.81it/s]
 79%|███████▉  | 27/34 [00:14<00:03,  1.81it/s]
 82%|████████▏ | 28/34 [00:15<00:03,  1.81it/s]
 85%|████████▌ | 29/34 [00:15<00:02,  1.81it/s]
 88%|████████▊ | 30/34 [00:16<00:02,  1.81it/s]
 91%|█████████ | 31/34 [00:17<00:01,  1.81it/s]
 94%|█████████▍| 32/34 [00:17<00:01,  1.81it/s]
 97%|█████████▋| 33/34 [00:18<00:00,  1.81it/s]
100%|██████████| 34/34 [00:18<00:00,  1.81it/s]
100%|██████████| 34/34 [00:18<00:00,  1.82it/s]
Total safe images: 2 out of 2

Version Details

Version ID: 8a23191c45a44cc45825733875e4409e462ec04f79d44abeac5ad50936f4e124
Version Created: November 5, 2025

Run on Replicate →