justmalhar/flux-thumbnails πΌοΈπ’βπβ β πΌοΈ
About
Generate 16:9 Thumbnails. Use prefix - `Thumbnail in the style of TOK`
Example Output
Prompt:
"Thumbnail in the style of TOK, A young woman with long black hair is posing in front of a green football field with glowing lights, displaying text "You wonβt believe thisβ¦β, She smiles at camera. The background features one to two thousand fans. In detail style. Realistic photography"
Output
Performance Metrics
25.06s
Prediction Time
25.09s
Total Time
All Input Parameters
{
"model": "dev",
"prompt": "Thumbnail in the style of TOK, A young woman with long black hair is posing in front of a green football field with glowing lights, displaying text \"You wonβt believe thisβ¦β, She smiles at camera. The background features one to two thousand fans. In detail style. Realistic photography",
"lora_scale": 1,
"num_outputs": 1,
"aspect_ratio": "16:9",
"output_format": "webp",
"guidance_scale": 3.5,
"output_quality": 100,
"num_inference_steps": 50
}
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
Using seed: 20163 Prompt: Thumbnail in the style of TOK, A young woman with long black hair is posing in front of a green football field with glowing lights, displaying text "You wonβt believe thisβ¦β, She smiles at camera. The background features one to two thousand fans. In detail style. Realistic photography txt2img mode Using dev model Loading LoRA weights Ensuring enough disk space... Free disk space: 9629617389568 Downloading weights 2024-08-21T02:35:12Z | INFO | [ Initiating ] chunk_size=150M dest=/src/weights-cache/97c19f42324058f7 url=https://replicate.delivery/yhqm/vLH3ZJVEeJ3SBanDU1Gp7nR3JZrYaIrLNMsIY7oD29jinXqJA/trained_model.tar 2024-08-21T02:35:15Z | INFO | [ Complete ] dest=/src/weights-cache/97c19f42324058f7 size="172 MB" total_elapsed=2.467s url=https://replicate.delivery/yhqm/vLH3ZJVEeJ3SBanDU1Gp7nR3JZrYaIrLNMsIY7oD29jinXqJA/trained_model.tar b'' Downloaded weights in 2.495424509048462 seconds LoRA weights loaded successfully 0%| | 0/50 [00:00<?, ?it/s] 2%|β | 1/50 [00:00<00:13, 3.67it/s] 4%|β | 2/50 [00:00<00:11, 4.24it/s] 6%|β | 3/50 [00:00<00:11, 3.97it/s] 8%|β | 4/50 [00:01<00:11, 3.85it/s] 10%|β | 5/50 [00:01<00:11, 3.78it/s] 12%|ββ | 6/50 [00:01<00:11, 3.75it/s] 14%|ββ | 7/50 [00:01<00:11, 3.73it/s] 16%|ββ | 8/50 [00:02<00:11, 3.71it/s] 18%|ββ | 9/50 [00:02<00:11, 3.70it/s] 20%|ββ | 10/50 [00:02<00:10, 3.70it/s] 22%|βββ | 11/50 [00:02<00:10, 3.70it/s] 24%|βββ | 12/50 [00:03<00:10, 3.68it/s] 26%|βββ | 13/50 [00:03<00:10, 3.68it/s] 28%|βββ | 14/50 [00:03<00:09, 3.69it/s] 30%|βββ | 15/50 [00:04<00:09, 3.69it/s] 32%|ββββ | 16/50 [00:04<00:09, 3.69it/s] 34%|ββββ | 17/50 [00:04<00:08, 3.68it/s] 36%|ββββ | 18/50 [00:04<00:08, 3.69it/s] 38%|ββββ | 19/50 [00:05<00:08, 3.69it/s] 40%|ββββ | 20/50 [00:05<00:08, 3.69it/s] 42%|βββββ | 21/50 [00:05<00:07, 3.68it/s] 44%|βββββ | 22/50 [00:05<00:07, 3.69it/s] 46%|βββββ | 23/50 [00:06<00:07, 3.69it/s] 48%|βββββ | 24/50 [00:06<00:07, 3.69it/s] 50%|βββββ | 25/50 [00:06<00:06, 3.68it/s] 52%|ββββββ | 26/50 [00:06<00:06, 3.69it/s] 54%|ββββββ | 27/50 [00:07<00:06, 3.69it/s] 56%|ββββββ | 28/50 [00:07<00:05, 3.68it/s] 58%|ββββββ | 29/50 [00:07<00:05, 3.68it/s] 60%|ββββββ | 30/50 [00:08<00:05, 3.69it/s] 62%|βββββββ | 31/50 [00:08<00:05, 3.69it/s] 64%|βββββββ | 32/50 [00:08<00:04, 3.68it/s] 66%|βββββββ | 33/50 [00:08<00:04, 3.68it/s] 68%|βββββββ | 34/50 [00:09<00:04, 3.69it/s] 70%|βββββββ | 35/50 [00:09<00:04, 3.69it/s] 72%|ββββββββ | 36/50 [00:09<00:03, 3.68it/s] 74%|ββββββββ | 37/50 [00:09<00:03, 3.68it/s] 76%|ββββββββ | 38/50 [00:10<00:03, 3.68it/s] 78%|ββββββββ | 39/50 [00:10<00:02, 3.69it/s] 80%|ββββββββ | 40/50 [00:10<00:02, 3.68it/s] 82%|βββββββββ | 41/50 [00:11<00:02, 3.68it/s] 84%|βββββββββ | 42/50 [00:11<00:02, 3.68it/s] 86%|βββββββββ | 43/50 [00:11<00:01, 3.69it/s] 88%|βββββββββ | 44/50 [00:11<00:01, 3.68it/s] 90%|βββββββββ | 45/50 [00:12<00:01, 3.68it/s] 92%|ββββββββββ| 46/50 [00:12<00:01, 3.68it/s] 94%|ββββββββββ| 47/50 [00:12<00:00, 3.69it/s] 96%|ββββββββββ| 48/50 [00:12<00:00, 3.68it/s] 98%|ββββββββββ| 49/50 [00:13<00:00, 3.68it/s] 100%|ββββββββββ| 50/50 [00:13<00:00, 3.69it/s] 100%|ββββββββββ| 50/50 [00:13<00:00, 3.70it/s]
Version Details
- Version ID
42799c2b58e0a6ca82d3a1d90f655f6386542e325d2017e3256b092189f567b8- Version Created
- August 21, 2024