hudsongraeme/cybertruck 🖼️🔢📝❓✓ → 🖼️

▶️ 538 runs 📅 Sep 2023 ⚙️ Cog 0.8.6

About

SDXL trained on a small cybertruck dataset

Example Output

Prompt:

"A photo of TOK drifting"

Output

Performance Metrics

61.77s Prediction Time

62.16s Total Time

All Input Parameters

{
  "width": 1024,
  "height": 1024,
  "prompt": "A photo of TOK drifting",
  "refine": "no_refiner",
  "scheduler": "K_EULER",
  "lora_scale": 0.6,
  "num_outputs": 4,
  "guidance_scale": 7.5,
  "apply_watermark": true,
  "high_noise_frac": 0.8,
  "negative_prompt": "",
  "prompt_strength": 0.8,
  "num_inference_steps": 50
}

Input Parameters

mask Type: string: Input mask for inpaint mode. Black areas will be preserved, white areas will be inpainted.
seed Type: integer: Random seed. Leave blank to randomize the seed
image Type: string: Input image for img2img or inpaint mode
width Type: integerDefault: 1024: Width of output image
height Type: integerDefault: 1024: Height of output image
prompt Type: stringDefault: An astronaut riding a rainbow unicorn: Input prompt
refine Default: no_refiner: Which refine style to use
scheduler Default: K_EULER: scheduler
lora_scale Type: numberDefault: 0.6Range: 0 - 1: LoRA additive scale. Only applicable on trained models.
num_outputs Type: integerDefault: 1Range: 1 - 4: Number of images to output.
refine_steps Type: integer: For base_image_refiner, the number of steps to refine, defaults to num_inference_steps
guidance_scale Type: numberDefault: 7.5Range: 1 - 50: Scale for classifier-free guidance
apply_watermark Type: booleanDefault: true: Applies a watermark to enable determining if an image is generated in downstream applications. If you have other provisions for generating or deploying images safely, you can use this to disable watermarking.
high_noise_frac Type: numberDefault: 0.8Range: 0 - 1: For expert_ensemble_refiner, the fraction of noise to use
negative_prompt Type: stringDefault:: Input Negative Prompt
prompt_strength Type: numberDefault: 0.8Range: 0 - 1: Prompt strength when using img2img / inpaint. 1.0 corresponds to full destruction of information in image
replicate_weights Type: string: Replicate LoRA weights to use. Leave blank to use the default weights.
num_inference_steps Type: integerDefault: 50Range: 1 - 500: Number of denoising steps
disable_safety_checker Type: booleanDefault: false: Disable safety checker for generated images. This feature is only available through the API. See https://replicate.com/docs/how-does-replicate-work#safety

Output Schema

Output

Type: array • Items Type: string • Items Format: uri

Example Execution Logs

Using seed: 880
Ensuring enough disk space...
Free disk space: 3320234479616
Downloading weights: https://pbxt.replicate.delivery/4h6fsYXIdRXfrkTavb4PfmCshSz7LnHFKsOyVMJXHSWa1kOjA/trained_model.tar
b''
Downloaded weights in 0.4663057327270508 seconds
Loading fine-tuned model
Does not have Unet. assume we are using LoRA
Loading Unet LoRA
Prompt: A photo of <s0><s1> drifting
txt2img mode
  0%|          | 0/50 [00:00<?, ?it/s]/root/.pyenv/versions/3.9.18/lib/python3.9/site-packages/torch/nn/modules/conv.py:459: UserWarning: Applied workaround for CuDNN issue, install nvrtc.so (Triggered internally at ../aten/src/ATen/native/cudnn/Conv_v8.cpp:80.)
return F.conv2d(input, weight, bias, self.stride,
  2%|▏         | 1/50 [00:01<01:21,  1.67s/it]
  4%|▍         | 2/50 [00:02<01:02,  1.31s/it]
  6%|▌         | 3/50 [00:03<00:56,  1.20s/it]
  8%|▊         | 4/50 [00:04<00:52,  1.15s/it]
 10%|█         | 5/50 [00:05<00:50,  1.12s/it]
 12%|█▏        | 6/50 [00:06<00:48,  1.10s/it]
 14%|█▍        | 7/50 [00:08<00:46,  1.09s/it]
 16%|█▌        | 8/50 [00:09<00:45,  1.08s/it]
 18%|█▊        | 9/50 [00:10<00:44,  1.08s/it]
 20%|██        | 10/50 [00:11<00:42,  1.07s/it]
 22%|██▏       | 11/50 [00:12<00:41,  1.07s/it]
 24%|██▍       | 12/50 [00:13<00:40,  1.07s/it]
 26%|██▌       | 13/50 [00:14<00:39,  1.06s/it]
 28%|██▊       | 14/50 [00:15<00:38,  1.06s/it]
 30%|███       | 15/50 [00:16<00:37,  1.06s/it]
 32%|███▏      | 16/50 [00:17<00:36,  1.06s/it]
 34%|███▍      | 17/50 [00:18<00:35,  1.06s/it]
 36%|███▌      | 18/50 [00:19<00:33,  1.06s/it]
 38%|███▊      | 19/50 [00:20<00:32,  1.06s/it]
 40%|████      | 20/50 [00:21<00:31,  1.06s/it]
 42%|████▏     | 21/50 [00:22<00:30,  1.06s/it]
 44%|████▍     | 22/50 [00:23<00:29,  1.06s/it]
 46%|████▌     | 23/50 [00:25<00:28,  1.06s/it]
 48%|████▊     | 24/50 [00:26<00:27,  1.06s/it]
 50%|█████     | 25/50 [00:27<00:26,  1.06s/it]
 52%|█████▏    | 26/50 [00:28<00:25,  1.07s/it]
 54%|█████▍    | 27/50 [00:29<00:24,  1.06s/it]
 56%|█████▌    | 28/50 [00:30<00:23,  1.07s/it]
 58%|█████▊    | 29/50 [00:31<00:22,  1.07s/it]
 60%|██████    | 30/50 [00:32<00:21,  1.07s/it]
 62%|██████▏   | 31/50 [00:33<00:20,  1.07s/it]
 64%|██████▍   | 32/50 [00:34<00:19,  1.07s/it]
 66%|██████▌   | 33/50 [00:35<00:18,  1.07s/it]
 68%|██████▊   | 34/50 [00:36<00:17,  1.07s/it]
 70%|███████   | 35/50 [00:37<00:16,  1.07s/it]
 72%|███████▏  | 36/50 [00:38<00:14,  1.07s/it]
 74%|███████▍  | 37/50 [00:40<00:13,  1.07s/it]
 76%|███████▌  | 38/50 [00:41<00:12,  1.07s/it]
 78%|███████▊  | 39/50 [00:42<00:11,  1.07s/it]
 80%|████████  | 40/50 [00:43<00:10,  1.07s/it]
 82%|████████▏ | 41/50 [00:44<00:09,  1.07s/it]
 84%|████████▍ | 42/50 [00:45<00:08,  1.07s/it]
 86%|████████▌ | 43/50 [00:46<00:07,  1.07s/it]
 88%|████████▊ | 44/50 [00:47<00:06,  1.07s/it]
 90%|█████████ | 45/50 [00:48<00:05,  1.07s/it]
 92%|█████████▏| 46/50 [00:49<00:04,  1.07s/it]
 94%|█████████▍| 47/50 [00:50<00:03,  1.07s/it]
 96%|█████████▌| 48/50 [00:51<00:02,  1.07s/it]
 98%|█████████▊| 49/50 [00:52<00:01,  1.07s/it]
100%|██████████| 50/50 [00:53<00:00,  1.07s/it]
100%|██████████| 50/50 [00:53<00:00,  1.08s/it]

Version Details

Version ID: 4e7b92920cf8bbec4862ccad2f905d83430d1ee54f47261d52e055aeadf6f9da
Version Created: October 24, 2023

Run on Replicate →