brockmesarich/thumbnailfinal 🖼️🔢❓📝✓ → 🖼️
About
Example Output
Prompt:
"
Create a headshot of @brock for use in YouTube thumbnails. The person is seated indoors in a modern, well-lit environment with soft, colorful lighting in the background (gradients of blue and pink). The subject wears a casual hoodie and looks directly at the camera with a confident and calm expression, featuring a slight closed-lip smile. The subject has a curious look on his face as if he is very curious about something. He is grabbing his chin as if he is thinking about something. The overall mood is professional, approachable, and composed. The framing focuses on the face, upper body, and gesture, ensuring the lighting and composition are visually appealing for thumbnails.
Make sure the skin on the face does not look too shiny
"Output
Performance Metrics
31.75s
Prediction Time
31.75s
Total Time
All Input Parameters
{
"model": "dev",
"prompt": "Create a headshot of @brock for use in YouTube thumbnails. The person is seated indoors in a modern, well-lit environment with soft, colorful lighting in the background (gradients of blue and pink). The subject wears a casual hoodie and looks directly at the camera with a confident and calm expression, featuring a slight closed-lip smile. The subject has a curious look on his face as if he is very curious about something. He is grabbing his chin as if he is thinking about something. The overall mood is professional, approachable, and composed. The framing focuses on the face, upper body, and gesture, ensuring the lighting and composition are visually appealing for thumbnails. \n\nMake sure the skin on the face does not look too shiny",
"go_fast": false,
"lora_scale": 1,
"megapixels": "1",
"num_outputs": 4,
"aspect_ratio": "16:9",
"output_format": "png",
"guidance_scale": 3,
"output_quality": 100,
"prompt_strength": 0.8,
"extra_lora_scale": 1,
"num_inference_steps": 28
}
Input Parameters
- mask
- Image mask for image inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
- seed
- Random seed. Set for reproducible generation
- image
- Input image for image to image or inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
- model
- 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
- 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
- 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)
- 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
- Run faster predictions with model optimized for speed (currently fp8 quantized); disable to run in original bf16
- extra_lora
- 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
- 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
- Approximate number of megapixels for generated image
- num_outputs
- Number of outputs to generate
- aspect_ratio
- Aspect ratio for the generated image. If custom is selected, uses height and width below & will run in bf16 mode
- output_format
- Format of the output images
- guidance_scale
- 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
- 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
- Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image
- extra_lora_scale
- 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
- 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
- Number of denoising steps. More steps can give more detailed images, but take longer.
- disable_safety_checker
- Disable safety checker for generated images.
Output Schema
Output
Example Execution Logs
free=26195492270080 Downloading weights 2025-04-19T15:55:14Z | INFO | [ Initiating ] chunk_size=150M dest=/tmp/tmpygn7lpik/weights url=https://replicate.delivery/xezq/c9hVAILYekQoVyFsVBfLGja1BfelELSypyAViyz0Bc47PscQB/trained_model.tar 2025-04-19T15:55:14Z | INFO | [ Cache Service ] enabled=true scheme=http target=hermes.services.svc.cluster.local 2025-04-19T15:55:16Z | INFO | [ Complete ] dest=/tmp/tmpygn7lpik/weights size="172 MB" total_elapsed=2.316s url=https://replicate.delivery/xezq/c9hVAILYekQoVyFsVBfLGja1BfelELSypyAViyz0Bc47PscQB/trained_model.tar Downloaded weights in 2.34s Loaded LoRAs in 2.94s Using seed: 4967 Prompt: Create a headshot of @brock for use in YouTube thumbnails. The person is seated indoors in a modern, well-lit environment with soft, colorful lighting in the background (gradients of blue and pink). The subject wears a casual hoodie and looks directly at the camera with a confident and calm expression, featuring a slight closed-lip smile. The subject has a curious look on his face as if he is very curious about something. He is grabbing his chin as if he is thinking about something. The overall mood is professional, approachable, and composed. The framing focuses on the face, upper body, and gesture, ensuring the lighting and composition are visually appealing for thumbnails. Make sure the skin on the face does not look too shiny [!] txt2img mode 0%| | 0/28 [00:00<?, ?it/s] 4%|▎ | 1/28 [00:00<00:26, 1.04it/s] 7%|▋ | 2/28 [00:01<00:22, 1.18it/s] 11%|█ | 3/28 [00:02<00:22, 1.11it/s] 14%|█▍ | 4/28 [00:03<00:22, 1.08it/s] 18%|█▊ | 5/28 [00:04<00:21, 1.06it/s] 21%|██▏ | 6/28 [00:05<00:20, 1.05it/s] 25%|██▌ | 7/28 [00:06<00:20, 1.05it/s] 29%|██▊ | 8/28 [00:07<00:19, 1.04it/s] 32%|███▏ | 9/28 [00:08<00:18, 1.04it/s] 36%|███▌ | 10/28 [00:09<00:17, 1.04it/s] 39%|███▉ | 11/28 [00:10<00:16, 1.04it/s] 43%|████▎ | 12/28 [00:11<00:15, 1.04it/s] 46%|████▋ | 13/28 [00:12<00:14, 1.04it/s] 50%|█████ | 14/28 [00:13<00:13, 1.04it/s] 54%|█████▎ | 15/28 [00:14<00:12, 1.04it/s] 57%|█████▋ | 16/28 [00:15<00:11, 1.04it/s] 61%|██████ | 17/28 [00:16<00:10, 1.04it/s] 64%|██████▍ | 18/28 [00:17<00:09, 1.04it/s] 68%|██████▊ | 19/28 [00:18<00:08, 1.04it/s] 71%|███████▏ | 20/28 [00:19<00:07, 1.04it/s] 75%|███████▌ | 21/28 [00:20<00:06, 1.04it/s] 79%|███████▊ | 22/28 [00:21<00:05, 1.04it/s] 82%|████████▏ | 23/28 [00:22<00:04, 1.04it/s] 86%|████████▌ | 24/28 [00:22<00:03, 1.04it/s] 89%|████████▉ | 25/28 [00:23<00:02, 1.04it/s] 93%|█████████▎| 26/28 [00:24<00:01, 1.04it/s] 96%|█████████▋| 27/28 [00:25<00:00, 1.04it/s] 100%|██████████| 28/28 [00:26<00:00, 1.04it/s] 100%|██████████| 28/28 [00:26<00:00, 1.04it/s] Total safe images: 4 out of 4
Version Details
- Version ID
5c553857be3a87f07df5825bc8eba031187166bb1e3d8841f5d5db3749d7223a- Version Created
- January 21, 2025