lucataco/realvisxl2-lora-inference 🖼️🔢📝❓✓ → 🖼️

▶️ 3.3K runs 📅 Nov 2023 ⚙️ Cog 0.8.5 🔗 GitHub ⚖️ License
image-inpainting image-to-image text-to-image

About

POC to run inference on Realvisxl2 LoRAs

Example Output

Prompt:

"A photo of TOK"

Output

Example output

Performance Metrics

25.24s Prediction Time
115.75s Total Time
All Input Parameters
{
  "seed": 6995,
  "width": 1024,
  "height": 1024,
  "prompt": "A photo of TOK",
  "refine": "no_refiner",
  "lora_url": "https://replicate.delivery/pbxt/L5zHkM0OHX4ZF1Ipnaiok6GHGvrRgZHBqbz2JjtBAtWz8mdE/trained_model.tar",
  "scheduler": "DPMSolverMultistep",
  "lora_scale": 0.6,
  "num_outputs": 1,
  "guidance_scale": 7.5,
  "apply_watermark": true,
  "high_noise_frac": 0.8,
  "negative_prompt": "(worst quality, low quality, illustration, 3d, 2d, painting, cartoons, sketch), open mouth",
  "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: A photo of TOK
Input prompt
refine Default: no_refiner
Which refine style to use
lora_url (required) Type: string
Load Lora model
scheduler Default: DPMSolverMultistep
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
num_inference_steps Type: integerDefault: 50Range: 1 - 500
Number of denoising steps
Output Schema

Output

Type: arrayItems Type: stringItems Format: uri

Example Execution Logs
LORA
Loading ssd txt2img pipeline...
Loading pipeline components...:   0%|          | 0/7 [00:00<?, ?it/s]
Loading pipeline components...:  29%|██▊       | 2/7 [00:00<00:00, 18.02it/s]
Loading pipeline components...:  57%|█████▋    | 4/7 [00:00<00:00, 12.25it/s]
Loading pipeline components...:  86%|████████▌ | 6/7 [00:00<00:00,  6.93it/s]
Loading pipeline components...: 100%|██████████| 7/7 [00:01<00:00,  5.83it/s]
Loading pipeline components...: 100%|██████████| 7/7 [00:01<00:00,  6.97it/s]
Loading ssd lora weights...
Loading fine-tuned model
Does not have Unet. Assume we are using LoRA
Loading Unet LoRA
Using seed: 6995
Prompt: A photo of <s0><s1>
txt2img mode
  0%|          | 0/50 [00:00<?, ?it/s]/root/.pyenv/versions/3.11.6/lib/python3.11/site-packages/diffusers/models/attention_processor.py:1815: FutureWarning: `LoRAAttnProcessor2_0` is deprecated and will be removed in version 0.26.0. Make sure use AttnProcessor2_0 instead by settingLoRA layers to `self.{to_q,to_k,to_v,to_out[0]}.lora_layer` respectively. This will be done automatically when using `LoraLoaderMixin.load_lora_weights`
deprecate(
  2%|▏         | 1/50 [00:00<00:18,  2.59it/s]
  4%|▍         | 2/50 [00:00<00:13,  3.65it/s]
  6%|▌         | 3/50 [00:00<00:12,  3.68it/s]
  8%|▊         | 4/50 [00:01<00:12,  3.70it/s]
 10%|█         | 5/50 [00:01<00:12,  3.70it/s]
 12%|█▏        | 6/50 [00:01<00:11,  3.70it/s]
 14%|█▍        | 7/50 [00:01<00:11,  3.71it/s]
 16%|█▌        | 8/50 [00:02<00:11,  3.71it/s]
 18%|█▊        | 9/50 [00:02<00:11,  3.71it/s]
 20%|██        | 10/50 [00:02<00:10,  3.70it/s]
 22%|██▏       | 11/50 [00:03<00:10,  3.71it/s]
 24%|██▍       | 12/50 [00:03<00:10,  3.70it/s]
 26%|██▌       | 13/50 [00:03<00:09,  3.71it/s]
 28%|██▊       | 14/50 [00:03<00:09,  3.71it/s]
 30%|███       | 15/50 [00:04<00:09,  3.71it/s]
 32%|███▏      | 16/50 [00:04<00:09,  3.70it/s]
 34%|███▍      | 17/50 [00:04<00:08,  3.70it/s]
 36%|███▌      | 18/50 [00:04<00:08,  3.70it/s]
 38%|███▊      | 19/50 [00:05<00:08,  3.70it/s]
 40%|████      | 20/50 [00:05<00:08,  3.70it/s]
 42%|████▏     | 21/50 [00:05<00:07,  3.70it/s]
 44%|████▍     | 22/50 [00:05<00:07,  3.70it/s]
 46%|████▌     | 23/50 [00:06<00:07,  3.70it/s]
 48%|████▊     | 24/50 [00:06<00:07,  3.70it/s]
 50%|█████     | 25/50 [00:06<00:06,  3.70it/s]
 52%|█████▏    | 26/50 [00:07<00:06,  3.70it/s]
 54%|█████▍    | 27/50 [00:07<00:06,  3.70it/s]
 56%|█████▌    | 28/50 [00:07<00:05,  3.70it/s]
 58%|█████▊    | 29/50 [00:07<00:05,  3.70it/s]
 60%|██████    | 30/50 [00:08<00:05,  3.70it/s]
 62%|██████▏   | 31/50 [00:08<00:05,  3.70it/s]
 64%|██████▍   | 32/50 [00:08<00:04,  3.70it/s]
 66%|██████▌   | 33/50 [00:08<00:04,  3.70it/s]
 68%|██████▊   | 34/50 [00:09<00:04,  3.70it/s]
 70%|███████   | 35/50 [00:09<00:04,  3.69it/s]
 72%|███████▏  | 36/50 [00:09<00:03,  3.69it/s]
 74%|███████▍  | 37/50 [00:10<00:03,  3.69it/s]
 76%|███████▌  | 38/50 [00:10<00:03,  3.69it/s]
 78%|███████▊  | 39/50 [00:10<00:02,  3.69it/s]
 80%|████████  | 40/50 [00:10<00:02,  3.69it/s]
 82%|████████▏ | 41/50 [00:11<00:02,  3.69it/s]
 84%|████████▍ | 42/50 [00:11<00:02,  3.69it/s]
 86%|████████▌ | 43/50 [00:11<00:01,  3.69it/s]
 88%|████████▊ | 44/50 [00:11<00:01,  3.69it/s]
 90%|█████████ | 45/50 [00:12<00:01,  3.69it/s]
 92%|█████████▏| 46/50 [00:12<00:01,  3.68it/s]
 94%|█████████▍| 47/50 [00:12<00:00,  3.68it/s]
 96%|█████████▌| 48/50 [00:13<00:00,  3.68it/s]
 98%|█████████▊| 49/50 [00:13<00:00,  3.68it/s]
100%|██████████| 50/50 [00:13<00:00,  3.68it/s]
100%|██████████| 50/50 [00:13<00:00,  3.69it/s]
Version Details
Version ID
9b5a0c77cd4f6bdb53a2c3d05b4774df02876d21dd7d37f13f518c03e996945b
Version Created
November 8, 2023
Run on Replicate →