fofr/sdxl-mass-effect-andromeda 🖼️🔢📝❓✓ → 🖼️
About
SDXL fine-tuned on Mass Effect Andromeda
Example Output
Prompt:
"A closeup TOK photo of an asari, unreal engine"
Output
Performance Metrics
7.58s
Prediction Time
10.53s
Total Time
All Input Parameters
{
"mask": null,
"seed": null,
"image": null,
"width": 1152,
"height": 768,
"prompt": "A closeup TOK photo of an asari, unreal engine",
"refine": "expert_ensemble_refiner",
"scheduler": "K_EULER",
"lora_scale": 0.8,
"num_outputs": 1,
"refine_steps": null,
"guidance_scale": 7.5,
"apply_watermark": false,
"high_noise_frac": 0.8,
"negative_prompt": "distorted, ugly, disfigured, broken",
"prompt_strength": 0.8,
"num_inference_steps": 30
}
Input Parameters
- mask
- Input mask for inpaint mode. Black areas will be preserved, white areas will be inpainted.
- seed
- Random seed. Leave blank to randomize the seed
- image
- Input image for img2img or inpaint mode
- width
- Width of output image
- height
- Height of output image
- prompt
- Input prompt
- refine
- Which refine style to use
- scheduler
- scheduler
- lora_scale
- LoRA additive scale. Only applicable on trained models.
- num_outputs
- Number of images to output.
- refine_steps
- For base_image_refiner, the number of steps to refine, defaults to num_inference_steps
- guidance_scale
- Scale for classifier-free guidance
- apply_watermark
- 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
- For expert_ensemble_refiner, the fraction of noise to use
- negative_prompt
- Input Negative Prompt
- prompt_strength
- Prompt strength when using img2img / inpaint. 1.0 corresponds to full destruction of information in image
- replicate_weights
- Replicate LoRA weights to use. Leave blank to use the default weights.
- num_inference_steps
- Number of denoising steps
- disable_safety_checker
- 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
Example Execution Logs
Using seed: 20685 Loading fine-tuned model Does not have Unet. assume we are using LoRA Loading Unet LoRA Prompt: A closeup <s0><s1> photo of an asari, unreal engine txt2img mode 0%| | 0/19 [00:00<?, ?it/s] 5%|▌ | 1/19 [00:00<00:04, 4.30it/s] 11%|█ | 2/19 [00:00<00:03, 4.29it/s] 16%|█▌ | 3/19 [00:00<00:03, 4.29it/s] 21%|██ | 4/19 [00:00<00:03, 4.30it/s] 26%|██▋ | 5/19 [00:01<00:03, 4.30it/s] 32%|███▏ | 6/19 [00:01<00:03, 4.30it/s] 37%|███▋ | 7/19 [00:01<00:02, 4.30it/s] 42%|████▏ | 8/19 [00:01<00:02, 4.29it/s] 47%|████▋ | 9/19 [00:02<00:02, 4.29it/s] 53%|█████▎ | 10/19 [00:02<00:02, 4.29it/s] 58%|█████▊ | 11/19 [00:02<00:01, 4.29it/s] 63%|██████▎ | 12/19 [00:02<00:01, 4.29it/s] 68%|██████▊ | 13/19 [00:03<00:01, 4.29it/s] 74%|███████▎ | 14/19 [00:03<00:01, 4.29it/s] 79%|███████▉ | 15/19 [00:03<00:00, 4.29it/s] 84%|████████▍ | 16/19 [00:03<00:00, 4.29it/s] 89%|████████▉ | 17/19 [00:03<00:00, 4.29it/s] 95%|█████████▍| 18/19 [00:04<00:00, 4.29it/s] 100%|██████████| 19/19 [00:04<00:00, 4.29it/s] 100%|██████████| 19/19 [00:04<00:00, 4.29it/s] 0%| | 0/7 [00:00<?, ?it/s] 14%|█▍ | 1/7 [00:00<00:01, 5.52it/s] 29%|██▊ | 2/7 [00:00<00:00, 5.51it/s] 43%|████▎ | 3/7 [00:00<00:00, 5.50it/s] 57%|█████▋ | 4/7 [00:00<00:00, 5.49it/s] 71%|███████▏ | 5/7 [00:00<00:00, 5.47it/s] 86%|████████▌ | 6/7 [00:01<00:00, 5.46it/s] 100%|██████████| 7/7 [00:01<00:00, 5.47it/s] 100%|██████████| 7/7 [00:01<00:00, 5.48it/s]
Version Details
- Version ID
565c07c44ae8e21077f7a895fe44daf0c471de72bb4c83abe6c588f83d063dce- Version Created
- October 30, 2023