okturan/flux-yesilcam ๐Ÿ–ผ๏ธ๐Ÿ”ขโ“๐Ÿ“โœ“ โ†’ ๐Ÿ–ผ๏ธ

โ–ถ๏ธ 174 runs ๐Ÿ“… Aug 2024 โš™๏ธ Cog 0.9.20
image-to-image lora text-to-image yesilcam

About

Creates images similar to scenes found in YeลŸilรงam movies

Example Output

Prompt:

"A scene from a YSLCM movie, A family in a station wagon parks two spots over from a grocer."

Output

Example outputExample outputExample outputExample output

Performance Metrics

48.65s Prediction Time
49.08s Total Time
All Input Parameters 
{
  "model": "dev",
  "width": 512,
  "height": 512,
  "prompt": "A scene from a YSLCM movie, A family in a station wagon parks two spots over from a grocer.",
  "lora_scale": 1.2,
  "num_outputs": 4,
  "aspect_ratio": "1:1",
  "output_format": "webp",
  "guidance_scale": 3,
  "output_quality": 80,
  "extra_lora_scale": 0.8,
  "num_inference_steps": 35
}
Input Parameters
mask Type: string
Image mask for image inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
seed Type: integer
Random seed. Set for reproducible generation
image Type: string
Input image for image to image or inpainting mode. If provided, aspect_ratio, width, and height inputs are ignored.
model Default: dev
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 Type: integerRange: 256 - 1440
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 Type: integerRange: 256 - 1440
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) Type: string
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 Type: booleanDefault: false
Run faster predictions with model optimized for speed (currently fp8 quantized); disable to run in original bf16
extra_lora Type: string
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 Type: numberDefault: 1Range: -1 - 3
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 Default: 1
Approximate number of megapixels for generated image
num_outputs Type: integerDefault: 1Range: 1 - 4
Number of outputs to generate
aspect_ratio Default: 1:1
Aspect ratio for the generated image. If custom is selected, uses height and width below & will run in bf16 mode
output_format Default: webp
Format of the output images
guidance_scale Type: numberDefault: 3Range: 0 - 10
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 Type: integerDefault: 80Range: 0 - 100
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 Type: numberDefault: 0.8Range: 0 - 1
Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image
extra_lora_scale Type: numberDefault: 1Range: -1 - 3
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 Type: string
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 Type: integerDefault: 28Range: 1 - 50
Number of denoising steps. More steps can give more detailed images, but take longer.
disable_safety_checker Type: booleanDefault: false
Disable safety checker for generated images.
Output Schema

Output

Type: array โ€ข Items Type: string โ€ข Items Format: uri

Example Execution Logs 
Using seed: 41750
Prompt: A scene from a YSLCM movie, A family in a station wagon parks two spots over from a grocer.
txt2img mode
Using dev model
free=9309006528512
Downloading weights
2024-08-28T22:54:27Z | INFO  | [ Initiating ] chunk_size=150M dest=/tmp/tmp4u0xbdbm/weights url=https://replicate.delivery/yhqm/9eHI5r9i1gRJB6KGX9J3Yw7o3fYkahHd0hfEfDfKf2ar4T11E/trained_model.tar
2024-08-28T22:54:30Z | INFO  | [ Complete ] dest=/tmp/tmp4u0xbdbm/weights size="172 MB" total_elapsed=3.289s url=https://replicate.delivery/yhqm/9eHI5r9i1gRJB6KGX9J3Yw7o3fYkahHd0hfEfDfKf2ar4T11E/trained_model.tar
Downloaded weights in 3.32s
Loaded LoRAs in 10.88s
  0%|          | 0/35 [00:00<?, ?it/s]
  3%|โ–Ž         | 1/35 [00:01<00:34,  1.02s/it]
  6%|โ–Œ         | 2/35 [00:01<00:29,  1.12it/s]
  9%|โ–Š         | 3/35 [00:02<00:30,  1.05it/s]
 11%|โ–ˆโ–        | 4/35 [00:03<00:30,  1.02it/s]
 14%|โ–ˆโ–        | 5/35 [00:04<00:29,  1.01it/s]
 17%|โ–ˆโ–‹        | 6/35 [00:05<00:29,  1.00s/it]
 20%|โ–ˆโ–ˆ        | 7/35 [00:06<00:28,  1.01s/it]
 23%|โ–ˆโ–ˆโ–Ž       | 8/35 [00:07<00:27,  1.01s/it]
 26%|โ–ˆโ–ˆโ–Œ       | 9/35 [00:08<00:26,  1.02s/it]
 29%|โ–ˆโ–ˆโ–Š       | 10/35 [00:09<00:25,  1.02s/it]
 31%|โ–ˆโ–ˆโ–ˆโ–      | 11/35 [00:11<00:24,  1.02s/it]
 34%|โ–ˆโ–ˆโ–ˆโ–      | 12/35 [00:12<00:23,  1.02s/it]
 37%|โ–ˆโ–ˆโ–ˆโ–‹      | 13/35 [00:13<00:22,  1.02s/it]
 40%|โ–ˆโ–ˆโ–ˆโ–ˆ      | 14/35 [00:14<00:21,  1.02s/it]
 43%|โ–ˆโ–ˆโ–ˆโ–ˆโ–Ž     | 15/35 [00:15<00:20,  1.02s/it]
 46%|โ–ˆโ–ˆโ–ˆโ–ˆโ–Œ     | 16/35 [00:16<00:19,  1.02s/it]
 49%|โ–ˆโ–ˆโ–ˆโ–ˆโ–Š     | 17/35 [00:17<00:18,  1.02s/it]
 51%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–    | 18/35 [00:18<00:17,  1.02s/it]
 54%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–    | 19/35 [00:19<00:16,  1.02s/it]
 57%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–‹    | 20/35 [00:20<00:15,  1.02s/it]
 60%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ    | 21/35 [00:21<00:14,  1.02s/it]
 63%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–Ž   | 22/35 [00:22<00:13,  1.02s/it]
 66%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–Œ   | 23/35 [00:23<00:12,  1.02s/it]
 69%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–Š   | 24/35 [00:24<00:11,  1.02s/it]
 71%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–  | 25/35 [00:25<00:10,  1.02s/it]
 74%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–  | 26/35 [00:26<00:09,  1.02s/it]
 77%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–‹  | 27/35 [00:27<00:08,  1.02s/it]
 80%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ  | 28/35 [00:28<00:07,  1.02s/it]
 83%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–Ž | 29/35 [00:29<00:06,  1.02s/it]
 86%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–Œ | 30/35 [00:30<00:05,  1.02s/it]
 89%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–Š | 31/35 [00:31<00:04,  1.02s/it]
 91%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–| 32/35 [00:32<00:03,  1.02s/it]
 94%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–| 33/35 [00:33<00:02,  1.02s/it]
 97%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–‹| 34/35 [00:34<00:01,  1.02s/it]
100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 35/35 [00:35<00:00,  1.02s/it]
100%|โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆ| 35/35 [00:35<00:00,  1.02s/it]
Version Details
Version ID
524103ba2ebc229fcc596d726b893449a77e987db29dc802629e1b6575244f07
Version Created
August 30, 2024
Run on Replicate โ†’