fermatresearch/open-dalle-1.1-lora 🖼️🔢📝❓✓ → 🖼️

▶️ 141.1K runs 📅 Dec 2023 ⚙️ Cog 0.8.6
image-inpainting image-to-image text-to-image

About

Better than SDXL at both prompt adherence and image quality, by dataautogpt3

Example Output

Prompt:

"shot in the style of sksfer, close up of a young woman in alaska, beautiful white hair, with a wolf necklace with a green gem in it, extreme detail and texture, professional photography"

Output

Example output

Performance Metrics

13.31s Prediction Time
13.34s Total Time
All Input Parameters 
{
  "seed": 54321,
  "width": 1024,
  "height": 1024,
  "prompt": "shot in the style of sksfer, close up of a young woman in alaska, beautiful white hair, with a wolf necklace with a green gem in it, extreme detail and texture, professional photography",
  "refine": "no_refiner",
  "scheduler": "K_EULER",
  "lora_scale": 0.65,
  "num_outputs": 1,
  "lora_weights": "https://replicate.delivery/pbxt/hM1H6f93HCVYQq471gZz6EYtRHPMJYAsyxeQXdGnozeDJKOkA/trained_model.tar",
  "guidance_scale": 7.5,
  "apply_watermark": false,
  "high_noise_frac": 0.8,
  "negative_prompt": "ugly, bad quality, nsfw",
  "prompt_strength": 0.8,
  "num_inference_steps": 35
}
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: KarrasDPM
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.
lora_weights Type: string
Replicate LoRA weights to use. Leave blank to use the default weights.
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: false
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: 35Range: 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](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: 54321
Loading fine-tuned model
Does not have Unet. assume we are using LoRA
Loading Unet LoRA
Prompt: shot in the style of <s0><s1>, close up of a young woman in alaska, beautiful white hair, with a wolf necklace with a green gem in it, extreme detail and texture, professional photography
txt2img mode
  0%|          | 0/35 [00:00<?, ?it/s]
  3%|▎         | 1/35 [00:00<00:09,  3.67it/s]
  6%|▌         | 2/35 [00:00<00:09,  3.66it/s]
  9%|▊         | 3/35 [00:00<00:08,  3.65it/s]
 11%|█▏        | 4/35 [00:01<00:08,  3.65it/s]
 14%|█▍        | 5/35 [00:01<00:08,  3.64it/s]
 17%|█▋        | 6/35 [00:01<00:07,  3.63it/s]
 20%|██        | 7/35 [00:01<00:07,  3.63it/s]
 23%|██▎       | 8/35 [00:02<00:07,  3.63it/s]
 26%|██▌       | 9/35 [00:02<00:07,  3.63it/s]
 29%|██▊       | 10/35 [00:02<00:06,  3.63it/s]
 31%|███▏      | 11/35 [00:03<00:06,  3.63it/s]
 34%|███▍      | 12/35 [00:03<00:06,  3.63it/s]
 37%|███▋      | 13/35 [00:03<00:06,  3.63it/s]
 40%|████      | 14/35 [00:03<00:05,  3.63it/s]
 43%|████▎     | 15/35 [00:04<00:05,  3.63it/s]
 46%|████▌     | 16/35 [00:04<00:05,  3.62it/s]
 49%|████▊     | 17/35 [00:04<00:04,  3.62it/s]
 51%|█████▏    | 18/35 [00:04<00:04,  3.62it/s]
 54%|█████▍    | 19/35 [00:05<00:04,  3.62it/s]
 57%|█████▋    | 20/35 [00:05<00:04,  3.62it/s]
 60%|██████    | 21/35 [00:05<00:03,  3.62it/s]
 63%|██████▎   | 22/35 [00:06<00:03,  3.62it/s]
 66%|██████▌   | 23/35 [00:06<00:03,  3.62it/s]
 69%|██████▊   | 24/35 [00:06<00:03,  3.62it/s]
 71%|███████▏  | 25/35 [00:06<00:02,  3.62it/s]
 74%|███████▍  | 26/35 [00:07<00:02,  3.62it/s]
 77%|███████▋  | 27/35 [00:07<00:02,  3.62it/s]
 80%|████████  | 28/35 [00:07<00:01,  3.62it/s]
 83%|████████▎ | 29/35 [00:07<00:01,  3.62it/s]
 86%|████████▌ | 30/35 [00:08<00:01,  3.62it/s]
 89%|████████▊ | 31/35 [00:08<00:01,  3.62it/s]
 91%|█████████▏| 32/35 [00:08<00:00,  3.62it/s]
 94%|█████████▍| 33/35 [00:09<00:00,  3.62it/s]
 97%|█████████▋| 34/35 [00:09<00:00,  3.62it/s]
100%|██████████| 35/35 [00:09<00:00,  3.62it/s]
100%|██████████| 35/35 [00:09<00:00,  3.62it/s]
Version Details
Version ID
2ade2cbfc88298b98366a6e361559e11666c17ed415d341c9ae776b30a61b196
Version Created
December 29, 2023
Run on Replicate →