philz1337x
/
clarity-upscaler
High resolution image Upscaler and Enhancer. Use at ClarityAI.co. A free Magnific alternative. Twitter/X: @philz1337x
Prediction
philz1337x/clarity-upscaler:dfad4170IDp6t3ufdbfkjp33cttiqxnspxpuStatusSucceededSourceWebHardwareA40Total durationCreatedInput
- seed
- 1337
- image
- prompt
- masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>
- dynamic
- 6
- scheduler
- DPM++ 3M SDE Karras
- creativity
- 0.35
- resemblance
- 0.6
- scale_factor
- 2
- negative_prompt
- (worst quality, low quality, normal quality:2) JuggernautNegative-neg
- num_inference_steps
- 18
{ "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVLkWEmioXxDchkSMf2srnR2gQuezEfWrW9NeKLpCeHcl5w/8_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 2, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 }npm install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import and set up the clientimport Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", { input: { seed: 1337, image: "https://replicate.delivery/pbxt/KZVLkWEmioXxDchkSMf2srnR2gQuezEfWrW9NeKLpCeHcl5w/8_before.png", prompt: "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", dynamic: 6, scheduler: "DPM++ 3M SDE Karras", creativity: 0.35, resemblance: 0.6, scale_factor: 2, negative_prompt: "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", num_inference_steps: 18 } } ); console.log(output);To learn more, take a look at the guide on getting started with Node.js.
pip install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import the clientimport replicateRun philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", input={ "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVLkWEmioXxDchkSMf2srnR2gQuezEfWrW9NeKLpCeHcl5w/8_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 2, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 } ) print(output)To learn more, take a look at the guide on getting started with Python.
Set theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVLkWEmioXxDchkSMf2srnR2gQuezEfWrW9NeKLpCeHcl5w/8_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 2, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 } }' \ https://api.replicate.com/v1/predictionsTo learn more, take a look at Replicate’s HTTP API reference docs.
Output
Loading...
{ "completed_at": "2024-03-15T02:52:35.038799Z", "created_at": "2024-03-15T02:52:09.801424Z", "data_removed": false, "error": null, "id": "p6t3ufdbfkjp33cttiqxnspxpu", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVLkWEmioXxDchkSMf2srnR2gQuezEfWrW9NeKLpCeHcl5w/8_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 2, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 }, "logs": "[Tiled Diffusion] upscaling image with 4x-UltraSharp...\n[Tiled Diffusion] ControlNet found, support is enabled.\n2024-03-15 02:52:12,365 - ControlNet - \u001b[0;32mINFO\u001b[0m - unit_separate = False, style_align = False\n2024-03-15 02:52:12,365 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loading model from cache: control_v11f1e_sd15_tile\n2024-03-15 02:52:12,377 - ControlNet - \u001b[0;32mINFO\u001b[0m - Using preprocessor: tile_resample\n2024-03-15 02:52:12,377 - ControlNet - \u001b[0;32mINFO\u001b[0m - preprocessor resolution = 1024\n2024-03-15 02:52:12,433 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet Hooked - Time = 0.07321739196777344\nMultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 4, Batch size: 4, Tile batches: 1 (ext: ContrlNet)\n[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling: 0%| | 0/7 [00:00<?, ?it/s]\n 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\nTotal progress: 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\n 14%|█▍ | 1/7 [00:02<00:16, 2.74s/it]\u001b[A\nTotal progress: 29%|██▊ | 2/7 [00:02<00:06, 1.37s/it]\u001b[A\n 29%|██▊ | 2/7 [00:05<00:13, 2.73s/it]\u001b[A\nTotal progress: 43%|████▎ | 3/7 [00:05<00:07, 1.94s/it]\u001b[A\n 43%|████▎ | 3/7 [00:08<00:10, 2.73s/it]\u001b[A\nTotal progress: 57%|█████▋ | 4/7 [00:08<00:06, 2.23s/it]\u001b[A\n 57%|█████▋ | 4/7 [00:10<00:08, 2.73s/it]\u001b[A\nTotal progress: 71%|███████▏ | 5/7 [00:10<00:04, 2.41s/it]\u001b[A\n 71%|███████▏ | 5/7 [00:13<00:05, 2.73s/it]\u001b[A\nTotal progress: 86%|████████▌ | 6/7 [00:13<00:02, 2.52s/it]\u001b[A\n 86%|████████▌ | 6/7 [00:16<00:02, 2.73s/it]\u001b[A\n100%|██████████| 7/7 [00:19<00:00, 2.73s/it]\u001b[A\n100%|██████████| 7/7 [00:19<00:00, 2.73s/it]\nTotal progress: 100%|██████████| 7/7 [00:16<00:00, 2.58s/it]\u001b[A[Tiled VAE]: the input size is tiny and unnecessary to tile.\nTotal progress: 100%|██████████| 7/7 [00:17<00:00, 2.58s/it]\u001b[A\nTotal progress: 100%|██████████| 7/7 [00:17<00:00, 2.44s/it]", "metrics": { "predict_time": 25.199207, "total_time": 25.237375 }, "output": [ "https://replicate.delivery/pbxt/t1kcfBTSChQfBUcGzXEVX7y95liS98WU2UNm8p9THuejPtAlA/1337-1257d396-e277-11ee-9c8c-2a0a4d26c304.png" ], "started_at": "2024-03-15T02:52:09.839592Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/p6t3ufdbfkjp33cttiqxnspxpu", "cancel": "https://api.replicate.com/v1/predictions/p6t3ufdbfkjp33cttiqxnspxpu/cancel" }, "version": "96c34bbe9aae48023bb102b0386f62a88ecd05bcdac34e95ca10857af055e895" }Generated in[Tiled Diffusion] upscaling image with 4x-UltraSharp... [Tiled Diffusion] ControlNet found, support is enabled. 2024-03-15 02:52:12,365 - ControlNet - INFO - unit_separate = False, style_align = False 2024-03-15 02:52:12,365 - ControlNet - INFO - Loading model from cache: control_v11f1e_sd15_tile 2024-03-15 02:52:12,377 - ControlNet - INFO - Using preprocessor: tile_resample 2024-03-15 02:52:12,377 - ControlNet - INFO - preprocessor resolution = 1024 2024-03-15 02:52:12,433 - ControlNet - INFO - ControlNet Hooked - Time = 0.07321739196777344 MultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 4, Batch size: 4, Tile batches: 1 (ext: ContrlNet) [Tiled VAE]: the input size is tiny and unnecessary to tile. MultiDiffusion Sampling: 0%| | 0/7 [00:00<?, ?it/s] 0%| | 0/7 [00:00<?, ?it/s] Total progress: 0%| | 0/7 [00:00<?, ?it/s] 14%|█▍ | 1/7 [00:02<00:16, 2.74s/it] Total progress: 29%|██▊ | 2/7 [00:02<00:06, 1.37s/it] 29%|██▊ | 2/7 [00:05<00:13, 2.73s/it] Total progress: 43%|████▎ | 3/7 [00:05<00:07, 1.94s/it] 43%|████▎ | 3/7 [00:08<00:10, 2.73s/it] Total progress: 57%|█████▋ | 4/7 [00:08<00:06, 2.23s/it] 57%|█████▋ | 4/7 [00:10<00:08, 2.73s/it] Total progress: 71%|███████▏ | 5/7 [00:10<00:04, 2.41s/it] 71%|███████▏ | 5/7 [00:13<00:05, 2.73s/it] Total progress: 86%|████████▌ | 6/7 [00:13<00:02, 2.52s/it] 86%|████████▌ | 6/7 [00:16<00:02, 2.73s/it] 100%|██████████| 7/7 [00:19<00:00, 2.73s/it] 100%|██████████| 7/7 [00:19<00:00, 2.73s/it] Total progress: 100%|██████████| 7/7 [00:16<00:00, 2.58s/it][Tiled VAE]: the input size is tiny and unnecessary to tile. Total progress: 100%|██████████| 7/7 [00:17<00:00, 2.58s/it] Total progress: 100%|██████████| 7/7 [00:17<00:00, 2.44s/it]Prediction
philz1337x/clarity-upscaler:dfad4170ID2mrrdt3btb2sufnpxvy4ngy2e4StatusSucceededSourceWebHardwareA100 (80GB)Total durationCreatedInput
- seed
- 1337
- image
- prompt
- (masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.2), extreme detailed,(joshua middleton comic cover art:1.1), (Action painting:1.2),(concretism:1.2), bird:1.3, leaves:1.8 ,(hypermaximalistic),colorful,highest detailed,<lora:78018:1>
- dynamic
- 5.55
- sd_model
- juggernaut_reborn.safetensors [338b85bc4f]
- scheduler
- DPM++ 3M SDE Karras
- creativity
- 0.51
- lora_links
- https://civitai.com/api/download/models/78018
- downscaling
- resemblance
- 1.3
- scale_factor
- 2
- tiling_width
- 144
- tiling_height
- 160
- negative_prompt
- (worst quality, low quality:2) face, person, woman, multiple heads multiple eyes
- num_inference_steps
- 18
- downscaling_resolution
- 768
{ "seed": 1337, "image": "https://replicate.delivery/pbxt/KdVUdf3TnhriFbyq5ooQn3DEEp63HuuO3RtgS7eb0jJHnLts/13_before.png", "prompt": "(masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.2), extreme detailed,(joshua middleton comic cover art:1.1), (Action painting:1.2),(concretism:1.2), bird:1.3, leaves:1.8 ,(hypermaximalistic),colorful,highest detailed,<lora:78018:1>", "dynamic": 5.55, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.51, "lora_links": "https://civitai.com/api/download/models/78018", "downscaling": false, "resemblance": 1.3, "scale_factor": 2, "tiling_width": 144, "tiling_height": 160, "negative_prompt": "(worst quality, low quality:2) face, person, woman, multiple heads multiple eyes", "num_inference_steps": 18, "downscaling_resolution": 768 }npm install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import and set up the clientimport Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", { input: { seed: 1337, image: "https://replicate.delivery/pbxt/KdVUdf3TnhriFbyq5ooQn3DEEp63HuuO3RtgS7eb0jJHnLts/13_before.png", prompt: "(masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.2), extreme detailed,(joshua middleton comic cover art:1.1), (Action painting:1.2),(concretism:1.2), bird:1.3, leaves:1.8 ,(hypermaximalistic),colorful,highest detailed,<lora:78018:1>", dynamic: 5.55, sd_model: "juggernaut_reborn.safetensors [338b85bc4f]", scheduler: "DPM++ 3M SDE Karras", creativity: 0.51, lora_links: "https://civitai.com/api/download/models/78018", downscaling: false, resemblance: 1.3, scale_factor: 2, tiling_width: 144, tiling_height: 160, negative_prompt: "(worst quality, low quality:2) face, person, woman, multiple heads multiple eyes", num_inference_steps: 18, downscaling_resolution: 768 } } ); console.log(output);To learn more, take a look at the guide on getting started with Node.js.
pip install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import the clientimport replicateRun philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", input={ "seed": 1337, "image": "https://replicate.delivery/pbxt/KdVUdf3TnhriFbyq5ooQn3DEEp63HuuO3RtgS7eb0jJHnLts/13_before.png", "prompt": "(masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.2), extreme detailed,(joshua middleton comic cover art:1.1), (Action painting:1.2),(concretism:1.2), bird:1.3, leaves:1.8 ,(hypermaximalistic),colorful,highest detailed,<lora:78018:1>", "dynamic": 5.55, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.51, "lora_links": "https://civitai.com/api/download/models/78018", "downscaling": False, "resemblance": 1.3, "scale_factor": 2, "tiling_width": 144, "tiling_height": 160, "negative_prompt": "(worst quality, low quality:2) face, person, woman, multiple heads multiple eyes", "num_inference_steps": 18, "downscaling_resolution": 768 } ) print(output)To learn more, take a look at the guide on getting started with Python.
Set theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KdVUdf3TnhriFbyq5ooQn3DEEp63HuuO3RtgS7eb0jJHnLts/13_before.png", "prompt": "(masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.2), extreme detailed,(joshua middleton comic cover art:1.1), (Action painting:1.2),(concretism:1.2), bird:1.3, leaves:1.8 ,(hypermaximalistic),colorful,highest detailed,<lora:78018:1>", "dynamic": 5.55, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.51, "lora_links": "https://civitai.com/api/download/models/78018", "downscaling": false, "resemblance": 1.3, "scale_factor": 2, "tiling_width": 144, "tiling_height": 160, "negative_prompt": "(worst quality, low quality:2) face, person, woman, multiple heads multiple eyes", "num_inference_steps": 18, "downscaling_resolution": 768 } }' \ https://api.replicate.com/v1/predictionsTo learn more, take a look at Replicate’s HTTP API reference docs.
Output
Loading...
{ "completed_at": "2024-03-26T10:04:32.800373Z", "created_at": "2024-03-26T10:04:01.234482Z", "data_removed": false, "error": null, "id": "2mrrdt3btb2sufnpxvy4ngy2e4", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KdVUdf3TnhriFbyq5ooQn3DEEp63HuuO3RtgS7eb0jJHnLts/13_before.png", "prompt": "(masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.2), extreme detailed,(joshua middleton comic cover art:1.1), (Action painting:1.2),(concretism:1.2), bird:1.3, leaves:1.8 ,(hypermaximalistic),colorful,highest detailed,<lora:78018:1>", "dynamic": 5.55, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.51, "lora_links": "https://civitai.com/api/download/models/78018", "downscaling": false, "resemblance": 1.3, "scale_factor": 2, "tiling_width": 144, "tiling_height": 160, "negative_prompt": "(worst quality, low quality:2) face, person, woman, multiple heads multiple eyes", "num_inference_steps": 18, "downscaling_resolution": 768 }, "logs": "Lora saved under: models/Lora/78018.safetensors\n[Tiled Diffusion] upscaling image with 4x-UltraSharp...\n[Tiled Diffusion] ControlNet found, support is enabled.\n2024-03-26 10:04:05,011 - ControlNet - \u001b[0;32mINFO\u001b[0m - unit_separate = False, style_align = False\n2024-03-26 10:04:05,012 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loading model from cache: control_v11f1e_sd15_tile\n2024-03-26 10:04:05,035 - ControlNet - \u001b[0;32mINFO\u001b[0m - Using preprocessor: tile_resample\n2024-03-26 10:04:05,035 - ControlNet - \u001b[0;32mINFO\u001b[0m - preprocessor resolution = 1536\n2024-03-26 10:04:05,113 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet Hooked - Time = 0.10718607902526855\nMultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 160x144, Tile count: 4, Batch size: 4, Tile batches: 1 (ext: ContrlNet)\n[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling: 0%| | 0/10 [00:00<?, ?it/s]\n 0%| | 0/10 [00:00<?, ?it/s]\u001b[A\u001b[A\nTotal progress: 0%| | 0/10 [00:00<?, ?it/s]\u001b[A\n 10%|█ | 1/10 [00:02<00:24, 2.75s/it]\u001b[A\u001b[A\nTotal progress: 20%|██ | 2/10 [00:02<00:09, 1.14s/it]\u001b[A\n 20%|██ | 2/10 [00:05<00:19, 2.48s/it]\u001b[A\u001b[A\nTotal progress: 30%|███ | 3/10 [00:04<00:11, 1.62s/it]\u001b[A\n 30%|███ | 3/10 [00:07<00:16, 2.39s/it]\u001b[A\u001b[A\nTotal progress: 40%|████ | 4/10 [00:06<00:11, 1.87s/it]\u001b[A\n 40%|████ | 4/10 [00:09<00:14, 2.35s/it]\u001b[A\u001b[A\nTotal progress: 50%|█████ | 5/10 [00:09<00:10, 2.01s/it]\u001b[A\n 50%|█████ | 5/10 [00:11<00:11, 2.33s/it]\u001b[A\u001b[A\nTotal progress: 60%|██████ | 6/10 [00:11<00:08, 2.10s/it]\u001b[A\n 60%|██████ | 6/10 [00:14<00:09, 2.31s/it]\u001b[A\u001b[A\nTotal progress: 70%|███████ | 7/10 [00:13<00:06, 2.16s/it]\u001b[A\n 70%|███████ | 7/10 [00:16<00:06, 2.30s/it]\u001b[A\u001b[A\nTotal progress: 80%|████████ | 8/10 [00:16<00:04, 2.20s/it]\u001b[A\n 80%|████████ | 8/10 [00:18<00:04, 2.30s/it]\u001b[A\u001b[A\nTotal progress: 90%|█████████ | 9/10 [00:18<00:02, 2.23s/it]\u001b[A\n 90%|█████████ | 9/10 [00:21<00:02, 2.29s/it]\u001b[A\u001b[A\n100%|██████████| 10/10 [00:23<00:00, 2.29s/it]\u001b[A\u001b[A\n100%|██████████| 10/10 [00:23<00:00, 2.33s/it]\nTotal progress: 100%|██████████| 10/10 [00:20<00:00, 2.25s/it]\u001b[A[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling: 45%|████▌ | 9/20 [01:08<01:23, 7.58s/it]\nTotal progress: 100%|██████████| 10/10 [00:21<00:00, 2.25s/it]\u001b[A\nTotal progress: 100%|██████████| 10/10 [00:21<00:00, 2.13s/it]", "metrics": { "predict_time": 31.516466, "total_time": 31.565891 }, "output": [ "https://replicate.delivery/pbxt/Z5K07xetYN24BaBeDe0IrerfbaYO9kWxiK4HIJKXUNx51ngUC/1337-3d0787aa-eb58-11ee-8ee1-e61d9bf0316a.png" ], "started_at": "2024-03-26T10:04:01.283907Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/2mrrdt3btb2sufnpxvy4ngy2e4", "cancel": "https://api.replicate.com/v1/predictions/2mrrdt3btb2sufnpxvy4ngy2e4/cancel" }, "version": "99c3e8b7d14d698e6c5485143c3c3fdf92d3c5e80e9c8f1b76b4ad41a0325a14" }Generated inLora saved under: models/Lora/78018.safetensors [Tiled Diffusion] upscaling image with 4x-UltraSharp... [Tiled Diffusion] ControlNet found, support is enabled. 2024-03-26 10:04:05,011 - ControlNet - INFO - unit_separate = False, style_align = False 2024-03-26 10:04:05,012 - ControlNet - INFO - Loading model from cache: control_v11f1e_sd15_tile 2024-03-26 10:04:05,035 - ControlNet - INFO - Using preprocessor: tile_resample 2024-03-26 10:04:05,035 - ControlNet - INFO - preprocessor resolution = 1536 2024-03-26 10:04:05,113 - ControlNet - INFO - ControlNet Hooked - Time = 0.10718607902526855 MultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 160x144, Tile count: 4, Batch size: 4, Tile batches: 1 (ext: ContrlNet) [Tiled VAE]: the input size is tiny and unnecessary to tile. MultiDiffusion Sampling: 0%| | 0/10 [00:00<?, ?it/s] 0%| | 0/10 [00:00<?, ?it/s] Total progress: 0%| | 0/10 [00:00<?, ?it/s] 10%|█ | 1/10 [00:02<00:24, 2.75s/it] Total progress: 20%|██ | 2/10 [00:02<00:09, 1.14s/it] 20%|██ | 2/10 [00:05<00:19, 2.48s/it] Total progress: 30%|███ | 3/10 [00:04<00:11, 1.62s/it] 30%|███ | 3/10 [00:07<00:16, 2.39s/it] Total progress: 40%|████ | 4/10 [00:06<00:11, 1.87s/it] 40%|████ | 4/10 [00:09<00:14, 2.35s/it] Total progress: 50%|█████ | 5/10 [00:09<00:10, 2.01s/it] 50%|█████ | 5/10 [00:11<00:11, 2.33s/it] Total progress: 60%|██████ | 6/10 [00:11<00:08, 2.10s/it] 60%|██████ | 6/10 [00:14<00:09, 2.31s/it] Total progress: 70%|███████ | 7/10 [00:13<00:06, 2.16s/it] 70%|███████ | 7/10 [00:16<00:06, 2.30s/it] Total progress: 80%|████████ | 8/10 [00:16<00:04, 2.20s/it] 80%|████████ | 8/10 [00:18<00:04, 2.30s/it] Total progress: 90%|█████████ | 9/10 [00:18<00:02, 2.23s/it] 90%|█████████ | 9/10 [00:21<00:02, 2.29s/it] 100%|██████████| 10/10 [00:23<00:00, 2.29s/it] 100%|██████████| 10/10 [00:23<00:00, 2.33s/it] Total progress: 100%|██████████| 10/10 [00:20<00:00, 2.25s/it][Tiled VAE]: the input size is tiny and unnecessary to tile. MultiDiffusion Sampling: 45%|████▌ | 9/20 [01:08<01:23, 7.58s/it] Total progress: 100%|██████████| 10/10 [00:21<00:00, 2.25s/it] Total progress: 100%|██████████| 10/10 [00:21<00:00, 2.13s/it]Prediction
philz1337x/clarity-upscaler:dfad4170IDwu337dtb5n4jdvygiy226r6uh4StatusSucceededSourceWebHardwareA100 (80GB)Total durationCreatedInput
- seed
- 1337
- image
- prompt
- masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>
- dynamic
- 6
- sd_model
- juggernaut_reborn.safetensors [338b85bc4f]
- scheduler
- DPM++ 3M SDE Karras
- creativity
- 0.5
- lora_links
- downscaling
- resemblance
- 0.6
- scale_factor
- 2
- tiling_width
- 112
- tiling_height
- 144
- negative_prompt
- (worst quality, low quality, normal quality:2) JuggernautNegative-neg
- num_inference_steps
- 18
- downscaling_resolution
- 768
{ "seed": 1337, "image": "https://replicate.delivery/pbxt/KeonPY6ZDWWUBrpVmOJ3v18xIpSQ0OWmfLwJI4xhTwJH0ns0/13_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.5, "lora_links": "", "downscaling": false, "resemblance": 0.6, "scale_factor": 2, "tiling_width": 112, "tiling_height": 144, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 768 }npm install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import and set up the clientimport Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", { input: { seed: 1337, image: "https://replicate.delivery/pbxt/KeonPY6ZDWWUBrpVmOJ3v18xIpSQ0OWmfLwJI4xhTwJH0ns0/13_before.png", prompt: "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", dynamic: 6, sd_model: "juggernaut_reborn.safetensors [338b85bc4f]", scheduler: "DPM++ 3M SDE Karras", creativity: 0.5, lora_links: "", downscaling: false, resemblance: 0.6, scale_factor: 2, tiling_width: 112, tiling_height: 144, negative_prompt: "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", num_inference_steps: 18, downscaling_resolution: 768 } } ); console.log(output);To learn more, take a look at the guide on getting started with Node.js.
pip install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import the clientimport replicateRun philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", input={ "seed": 1337, "image": "https://replicate.delivery/pbxt/KeonPY6ZDWWUBrpVmOJ3v18xIpSQ0OWmfLwJI4xhTwJH0ns0/13_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.5, "lora_links": "", "downscaling": False, "resemblance": 0.6, "scale_factor": 2, "tiling_width": 112, "tiling_height": 144, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 768 } ) print(output)To learn more, take a look at the guide on getting started with Python.
Set theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KeonPY6ZDWWUBrpVmOJ3v18xIpSQ0OWmfLwJI4xhTwJH0ns0/13_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.5, "lora_links": "", "downscaling": false, "resemblance": 0.6, "scale_factor": 2, "tiling_width": 112, "tiling_height": 144, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 768 } }' \ https://api.replicate.com/v1/predictionsTo learn more, take a look at Replicate’s HTTP API reference docs.
Output
Loading...
{ "completed_at": "2024-03-30T02:09:02.889242Z", "created_at": "2024-03-30T02:08:45.060971Z", "data_removed": false, "error": null, "id": "wu337dtb5n4jdvygiy226r6uh4", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KeonPY6ZDWWUBrpVmOJ3v18xIpSQ0OWmfLwJI4xhTwJH0ns0/13_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.5, "lora_links": "", "downscaling": false, "resemblance": 0.6, "scale_factor": 2, "tiling_width": 112, "tiling_height": 144, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 768 }, "logs": "Running prediction\n[Tiled Diffusion] upscaling image with 4x-UltraSharp...\n[Tiled Diffusion] ControlNet found, support is enabled.\n2024-03-30 02:08:47,641 - ControlNet - \u001b[0;32mINFO\u001b[0m - unit_separate = False, style_align = False\n2024-03-30 02:08:47,641 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loading model from cache: control_v11f1e_sd15_tile\n2024-03-30 02:08:47,663 - ControlNet - \u001b[0;32mINFO\u001b[0m - Using preprocessor: tile_resample\n2024-03-30 02:08:47,663 - ControlNet - \u001b[0;32mINFO\u001b[0m - preprocessor resolution = 1536\n2024-03-30 02:08:47,736 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet Hooked - Time = 0.09977936744689941\nMultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 4, Batch size: 4, Tile batches: 1 (ext: ContrlNet)\n[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling: 0%| | 0/7 [00:00<?, ?it/s]\n 0%| | 0/10 [00:00<?, ?it/s]\u001b[A\u001b[A\nTotal progress: 0%| | 0/10 [00:00<?, ?it/s]\u001b[A\n 10%|█ | 1/10 [00:01<00:11, 1.28s/it]\u001b[A\u001b[A\nTotal progress: 20%|██ | 2/10 [00:01<00:05, 1.57it/s]\u001b[A\n 20%|██ | 2/10 [00:02<00:10, 1.27s/it]\u001b[A\u001b[A\nTotal progress: 30%|███ | 3/10 [00:02<00:06, 1.11it/s]\u001b[A\n 30%|███ | 3/10 [00:03<00:08, 1.27s/it]\u001b[A\u001b[A\nTotal progress: 40%|████ | 4/10 [00:03<00:06, 1.04s/it]\u001b[A\n 40%|████ | 4/10 [00:05<00:07, 1.27s/it]\u001b[A\u001b[A\nTotal progress: 50%|█████ | 5/10 [00:05<00:05, 1.12s/it]\u001b[A\n 50%|█████ | 5/10 [00:06<00:06, 1.27s/it]\u001b[A\u001b[A\nTotal progress: 60%|██████ | 6/10 [00:06<00:04, 1.17s/it]\u001b[A\n 60%|██████ | 6/10 [00:07<00:05, 1.27s/it]\u001b[A\u001b[A\nTotal progress: 70%|███████ | 7/10 [00:07<00:03, 1.20s/it]\u001b[A\n 70%|███████ | 7/10 [00:08<00:03, 1.27s/it]\u001b[A\u001b[A\nTotal progress: 80%|████████ | 8/10 [00:08<00:02, 1.22s/it]\u001b[A\n 80%|████████ | 8/10 [00:10<00:02, 1.27s/it]\u001b[A\u001b[A\nTotal progress: 90%|█████████ | 9/10 [00:10<00:01, 1.24s/it]\u001b[A\n 90%|█████████ | 9/10 [00:11<00:01, 1.27s/it]\u001b[A\u001b[A\n100%|██████████| 10/10 [00:12<00:00, 1.27s/it]\u001b[A\u001b[A\n100%|██████████| 10/10 [00:12<00:00, 1.27s/it]\nTotal progress: 100%|██████████| 10/10 [00:11<00:00, 1.25s/it]\u001b[A[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling: 0%| | 0/9 [01:04<?, ?it/s]\nTotal progress: 100%|██████████| 10/10 [00:12<00:00, 1.25s/it]\u001b[A\nTotal progress: 100%|██████████| 10/10 [00:12<00:00, 1.21s/it]", "metrics": { "predict_time": 17.807848, "total_time": 17.828271 }, "output": [ "https://replicate.delivery/pbxt/tUh8fJ7aZwRJN6DATcqhcS4BwwnDRz645kvpMF07gLIfYSlSA/1337-7a3c31cc-ee3a-11ee-9eea-facc40b28a6c.png" ], "started_at": "2024-03-30T02:08:45.081394Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/wu337dtb5n4jdvygiy226r6uh4", "cancel": "https://api.replicate.com/v1/predictions/wu337dtb5n4jdvygiy226r6uh4/cancel" }, "version": "c2d38005518d6610a3c8061c607e65d57e004a0fc985b3a2046e1a50f9f0d4fc" }Generated inRunning prediction [Tiled Diffusion] upscaling image with 4x-UltraSharp... [Tiled Diffusion] ControlNet found, support is enabled. 2024-03-30 02:08:47,641 - ControlNet - INFO - unit_separate = False, style_align = False 2024-03-30 02:08:47,641 - ControlNet - INFO - Loading model from cache: control_v11f1e_sd15_tile 2024-03-30 02:08:47,663 - ControlNet - INFO - Using preprocessor: tile_resample 2024-03-30 02:08:47,663 - ControlNet - INFO - preprocessor resolution = 1536 2024-03-30 02:08:47,736 - ControlNet - INFO - ControlNet Hooked - Time = 0.09977936744689941 MultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 4, Batch size: 4, Tile batches: 1 (ext: ContrlNet) [Tiled VAE]: the input size is tiny and unnecessary to tile. MultiDiffusion Sampling: 0%| | 0/7 [00:00<?, ?it/s] 0%| | 0/10 [00:00<?, ?it/s] Total progress: 0%| | 0/10 [00:00<?, ?it/s] 10%|█ | 1/10 [00:01<00:11, 1.28s/it] Total progress: 20%|██ | 2/10 [00:01<00:05, 1.57it/s] 20%|██ | 2/10 [00:02<00:10, 1.27s/it] Total progress: 30%|███ | 3/10 [00:02<00:06, 1.11it/s] 30%|███ | 3/10 [00:03<00:08, 1.27s/it] Total progress: 40%|████ | 4/10 [00:03<00:06, 1.04s/it] 40%|████ | 4/10 [00:05<00:07, 1.27s/it] Total progress: 50%|█████ | 5/10 [00:05<00:05, 1.12s/it] 50%|█████ | 5/10 [00:06<00:06, 1.27s/it] Total progress: 60%|██████ | 6/10 [00:06<00:04, 1.17s/it] 60%|██████ | 6/10 [00:07<00:05, 1.27s/it] Total progress: 70%|███████ | 7/10 [00:07<00:03, 1.20s/it] 70%|███████ | 7/10 [00:08<00:03, 1.27s/it] Total progress: 80%|████████ | 8/10 [00:08<00:02, 1.22s/it] 80%|████████ | 8/10 [00:10<00:02, 1.27s/it] Total progress: 90%|█████████ | 9/10 [00:10<00:01, 1.24s/it] 90%|█████████ | 9/10 [00:11<00:01, 1.27s/it] 100%|██████████| 10/10 [00:12<00:00, 1.27s/it] 100%|██████████| 10/10 [00:12<00:00, 1.27s/it] Total progress: 100%|██████████| 10/10 [00:11<00:00, 1.25s/it][Tiled VAE]: the input size is tiny and unnecessary to tile. MultiDiffusion Sampling: 0%| | 0/9 [01:04<?, ?it/s] Total progress: 100%|██████████| 10/10 [00:12<00:00, 1.25s/it] Total progress: 100%|██████████| 10/10 [00:12<00:00, 1.21s/it]Prediction
philz1337x/clarity-upscaler:dfad4170IDrz2tqztbaqls6k6ki5hcy2j5iyStatusSucceededSourceWebHardwareA40Total durationCreatedInput
- seed
- 1337
- image
- prompt
- masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>, black man, white shirt, in the street, sunrise, america
- dynamic
- 6
- sd_model
- juggernaut_reborn.safetensors [338b85bc4f]
- scheduler
- DPM++ 3M SDE Karras
- creativity
- 0.4
- downscaling
- resemblance
- 0.6
- scale_factor
- 4
- tiling_width
- 112
- tiling_height
- 144
- negative_prompt
- (worst quality, low quality, normal quality:2) JuggernautNegative-neg
- num_inference_steps
- 18
- downscaling_resolution
- 512
{ "seed": 1337, "image": "https://replicate.delivery/pbxt/KbfBTAyIUmJBWaWXXB7ZCVrWHomXCNfRWKyslpcJmzoyLZkt/ahshitherewegoagain.jpg", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>, black man, white shirt, in the street, sunrise, america", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.4, "downscaling": true, "resemblance": 0.6, "scale_factor": 4, "tiling_width": 112, "tiling_height": 144, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 512 }npm install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import and set up the clientimport Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", { input: { seed: 1337, image: "https://replicate.delivery/pbxt/KbfBTAyIUmJBWaWXXB7ZCVrWHomXCNfRWKyslpcJmzoyLZkt/ahshitherewegoagain.jpg", prompt: "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>, black man, white shirt, in the street, sunrise, america", dynamic: 6, sd_model: "juggernaut_reborn.safetensors [338b85bc4f]", scheduler: "DPM++ 3M SDE Karras", creativity: 0.4, downscaling: true, resemblance: 0.6, scale_factor: 4, tiling_width: 112, tiling_height: 144, negative_prompt: "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", num_inference_steps: 18, downscaling_resolution: 512 } } ); console.log(output);To learn more, take a look at the guide on getting started with Node.js.
pip install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import the clientimport replicateRun philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", input={ "seed": 1337, "image": "https://replicate.delivery/pbxt/KbfBTAyIUmJBWaWXXB7ZCVrWHomXCNfRWKyslpcJmzoyLZkt/ahshitherewegoagain.jpg", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>, black man, white shirt, in the street, sunrise, america", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.4, "downscaling": True, "resemblance": 0.6, "scale_factor": 4, "tiling_width": 112, "tiling_height": 144, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 512 } ) print(output)To learn more, take a look at the guide on getting started with Python.
Set theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KbfBTAyIUmJBWaWXXB7ZCVrWHomXCNfRWKyslpcJmzoyLZkt/ahshitherewegoagain.jpg", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>, black man, white shirt, in the street, sunrise, america", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.4, "downscaling": true, "resemblance": 0.6, "scale_factor": 4, "tiling_width": 112, "tiling_height": 144, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 512 } }' \ https://api.replicate.com/v1/predictionsTo learn more, take a look at Replicate’s HTTP API reference docs.
Output
Loading...
{ "completed_at": "2024-03-21T04:51:12.793037Z", "created_at": "2024-03-21T04:49:18.441158Z", "data_removed": false, "error": null, "id": "rz2tqztbaqls6k6ki5hcy2j5iy", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KbfBTAyIUmJBWaWXXB7ZCVrWHomXCNfRWKyslpcJmzoyLZkt/ahshitherewegoagain.jpg", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>, black man, white shirt, in the street, sunrise, america", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.4, "downscaling": true, "resemblance": 0.6, "scale_factor": 4, "tiling_width": 112, "tiling_height": 144, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 512 }, "logs": "[Tiled Diffusion] upscaling image with 4x-UltraSharp...\n[Tiled Diffusion] ControlNet found, support is enabled.\n2024-03-21 04:50:34,517 - ControlNet - \u001b[0;32mINFO\u001b[0m - unit_separate = False, style_align = False\n2024-03-21 04:50:34,517 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loading model from cache: control_v11f1e_sd15_tile\n2024-03-21 04:50:34,520 - ControlNet - \u001b[0;32mINFO\u001b[0m - Using preprocessor: tile_resample\n2024-03-21 04:50:34,520 - ControlNet - \u001b[0;32mINFO\u001b[0m - preprocessor resolution = 1364\n2024-03-21 04:50:34,589 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet Hooked - Time = 0.07781791687011719\nMultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 6, Batch size: 6, Tile batches: 1 (ext: ContrlNet)\n[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling: 0%| | 0/8 [00:00<?, ?it/s]\u001b[A\n 0%| | 0/8 [00:00<?, ?it/s]\u001b[A\u001b[A\nTotal progress: 0%| | 0/8 [00:00<?, ?it/s]\u001b[A\n 12%|█▎ | 1/8 [00:04<00:28, 4.12s/it]\u001b[A\u001b[A\nTotal progress: 25%|██▌ | 2/8 [00:04<00:12, 2.04s/it]\u001b[A\n 25%|██▌ | 2/8 [00:08<00:24, 4.10s/it]\u001b[A\u001b[A\nTotal progress: 38%|███▊ | 3/8 [00:08<00:14, 2.90s/it]\u001b[A\n 38%|███▊ | 3/8 [00:12<00:20, 4.09s/it]\u001b[A\u001b[A\nTotal progress: 50%|█████ | 4/8 [00:12<00:13, 3.34s/it]\u001b[A\n 50%|█████ | 4/8 [00:16<00:16, 4.09s/it]\u001b[A\u001b[A\nTotal progress: 62%|██████▎ | 5/8 [00:16<00:10, 3.60s/it]\u001b[A\n 62%|██████▎ | 5/8 [00:20<00:12, 4.09s/it]\u001b[A\u001b[A\nTotal progress: 75%|███████▌ | 6/8 [00:20<00:07, 3.76s/it]\u001b[A\n 75%|███████▌ | 6/8 [00:24<00:08, 4.09s/it]\u001b[A\u001b[A\nTotal progress: 88%|████████▊ | 7/8 [00:24<00:03, 3.87s/it]\u001b[A\n 88%|████████▊ | 7/8 [00:28<00:04, 4.09s/it]\u001b[A\u001b[A\n100%|██████████| 8/8 [00:32<00:00, 4.09s/it]\u001b[A\u001b[A\n100%|██████████| 8/8 [00:32<00:00, 4.09s/it]\nMultiDiffusion Sampling: 0%| | 0/8 [00:42<?, ?it/s]\nTotal progress: 100%|██████████| 8/8 [00:28<00:00, 3.94s/it]\u001b[A[Tiled VAE]: input_size: torch.Size([1, 4, 170, 256]), tile_size: 192, padding: 11\n[Tiled VAE]: split to 1x2 = 2 tiles. Optimal tile size 128x160, original tile size 192x192\n[Tiled VAE]: Fast mode enabled, estimating group norm parameters on 192 x 127 image\n[Tiled VAE]: Executing Decoder Task Queue: 0%| | 0/246 [00:00<?, ?it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 50%|█████ | 124/246 [00:00<00:00, 333.58it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 246/246 [00:00<00:00, 351.31it/s]\n[Tiled VAE]: Done in 1.736s, max VRAM alloc 6606.047 MB\nTotal progress: 100%|██████████| 8/8 [00:30<00:00, 3.94s/it]\u001b[A\nTotal progress: 100%|██████████| 8/8 [00:30<00:00, 3.84s/it]", "metrics": { "predict_time": 39.694987, "total_time": 114.351879 }, "output": [ "https://replicate.delivery/pbxt/TgHff4TNKbrvSEV5BgI3ubni5SLq878sGRWtNoeRY9CerbJKB/1337-a37afce2-e73e-11ee-9f62-7e41a39613fc.png" ], "started_at": "2024-03-21T04:50:33.098050Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/rz2tqztbaqls6k6ki5hcy2j5iy", "cancel": "https://api.replicate.com/v1/predictions/rz2tqztbaqls6k6ki5hcy2j5iy/cancel" }, "version": "abd484acb51ad450b06f42f76940fa5c1b37511dbf70ac8594fdacd5c3302307" }Generated in[Tiled Diffusion] upscaling image with 4x-UltraSharp... [Tiled Diffusion] ControlNet found, support is enabled. 2024-03-21 04:50:34,517 - ControlNet - INFO - unit_separate = False, style_align = False 2024-03-21 04:50:34,517 - ControlNet - INFO - Loading model from cache: control_v11f1e_sd15_tile 2024-03-21 04:50:34,520 - ControlNet - INFO - Using preprocessor: tile_resample 2024-03-21 04:50:34,520 - ControlNet - INFO - preprocessor resolution = 1364 2024-03-21 04:50:34,589 - ControlNet - INFO - ControlNet Hooked - Time = 0.07781791687011719 MultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 6, Batch size: 6, Tile batches: 1 (ext: ContrlNet) [Tiled VAE]: the input size is tiny and unnecessary to tile. MultiDiffusion Sampling: 0%| | 0/8 [00:00<?, ?it/s] 0%| | 0/8 [00:00<?, ?it/s] Total progress: 0%| | 0/8 [00:00<?, ?it/s] 12%|█▎ | 1/8 [00:04<00:28, 4.12s/it] Total progress: 25%|██▌ | 2/8 [00:04<00:12, 2.04s/it] 25%|██▌ | 2/8 [00:08<00:24, 4.10s/it] Total progress: 38%|███▊ | 3/8 [00:08<00:14, 2.90s/it] 38%|███▊ | 3/8 [00:12<00:20, 4.09s/it] Total progress: 50%|█████ | 4/8 [00:12<00:13, 3.34s/it] 50%|█████ | 4/8 [00:16<00:16, 4.09s/it] Total progress: 62%|██████▎ | 5/8 [00:16<00:10, 3.60s/it] 62%|██████▎ | 5/8 [00:20<00:12, 4.09s/it] Total progress: 75%|███████▌ | 6/8 [00:20<00:07, 3.76s/it] 75%|███████▌ | 6/8 [00:24<00:08, 4.09s/it] Total progress: 88%|████████▊ | 7/8 [00:24<00:03, 3.87s/it] 88%|████████▊ | 7/8 [00:28<00:04, 4.09s/it] 100%|██████████| 8/8 [00:32<00:00, 4.09s/it] 100%|██████████| 8/8 [00:32<00:00, 4.09s/it] MultiDiffusion Sampling: 0%| | 0/8 [00:42<?, ?it/s] Total progress: 100%|██████████| 8/8 [00:28<00:00, 3.94s/it][Tiled VAE]: input_size: torch.Size([1, 4, 170, 256]), tile_size: 192, padding: 11 [Tiled VAE]: split to 1x2 = 2 tiles. Optimal tile size 128x160, original tile size 192x192 [Tiled VAE]: Fast mode enabled, estimating group norm parameters on 192 x 127 image [Tiled VAE]: Executing Decoder Task Queue: 0%| | 0/246 [00:00<?, ?it/s] [Tiled VAE]: Executing Decoder Task Queue: 50%|█████ | 124/246 [00:00<00:00, 333.58it/s] [Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 246/246 [00:00<00:00, 351.31it/s] [Tiled VAE]: Done in 1.736s, max VRAM alloc 6606.047 MB Total progress: 100%|██████████| 8/8 [00:30<00:00, 3.94s/it] Total progress: 100%|██████████| 8/8 [00:30<00:00, 3.84s/it]Prediction
philz1337x/clarity-upscaler:dfad4170IDpojd5ydbfoozq3rtylgsidu6jmStatusSucceededSourceWebHardwareA40Total durationCreatedInput
- seed
- 1337
- image
- prompt
- masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>
- dynamic
- 6
- scheduler
- DPM++ 3M SDE Karras
- creativity
- 0.35
- resemblance
- 0.6
- scale_factor
- 4
- negative_prompt
- (worst quality, low quality, normal quality:2) JuggernautNegative-neg
- num_inference_steps
- 18
{ "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVNjQNv2EUrbLHAmEkCWGZQOmqgb462BkdKtwiaBbnhZbLZ/ai-4.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 }npm install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import and set up the clientimport Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", { input: { seed: 1337, image: "https://replicate.delivery/pbxt/KZVNjQNv2EUrbLHAmEkCWGZQOmqgb462BkdKtwiaBbnhZbLZ/ai-4.png", prompt: "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", dynamic: 6, scheduler: "DPM++ 3M SDE Karras", creativity: 0.35, resemblance: 0.6, scale_factor: 4, negative_prompt: "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", num_inference_steps: 18 } } ); console.log(output);To learn more, take a look at the guide on getting started with Node.js.
pip install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import the clientimport replicateRun philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", input={ "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVNjQNv2EUrbLHAmEkCWGZQOmqgb462BkdKtwiaBbnhZbLZ/ai-4.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 } ) print(output)To learn more, take a look at the guide on getting started with Python.
Set theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVNjQNv2EUrbLHAmEkCWGZQOmqgb462BkdKtwiaBbnhZbLZ/ai-4.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 } }' \ https://api.replicate.com/v1/predictionsTo learn more, take a look at Replicate’s HTTP API reference docs.
Output
Loading...
{ "completed_at": "2024-03-15T03:00:56.719485Z", "created_at": "2024-03-15T02:56:47.971365Z", "data_removed": false, "error": null, "id": "pojd5ydbfoozq3rtylgsidu6jm", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVNjQNv2EUrbLHAmEkCWGZQOmqgb462BkdKtwiaBbnhZbLZ/ai-4.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 }, "logs": "[Tiled Diffusion] upscaling image with 4x-UltraSharp...\n[Tiled Diffusion] ControlNet found, support is enabled.\n2024-03-15 03:00:05,196 - ControlNet - \u001b[0;32mINFO\u001b[0m - unit_separate = False, style_align = False\n2024-03-15 03:00:05,196 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loading model from cache: control_v11f1e_sd15_tile\n2024-03-15 03:00:05,208 - ControlNet - \u001b[0;32mINFO\u001b[0m - Using preprocessor: tile_resample\n2024-03-15 03:00:05,208 - ControlNet - \u001b[0;32mINFO\u001b[0m - preprocessor resolution = 2048\n2024-03-15 03:00:05,309 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet Hooked - Time = 0.11690354347229004\nMultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 9, Batch size: 5, Tile batches: 2 (ext: ContrlNet)\n[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling: 0%| | 0/14 [00:00<?, ?it/s]\u001b[A\n 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\u001b[A\nMultiDiffusion Sampling: 7%|▋ | 1/14 [00:03<00:45, 3.46s/it]\u001b[A\nTotal progress: 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\n 14%|█▍ | 1/7 [00:06<00:38, 6.37s/it]\u001b[A\u001b[A\n 29%|██▊ | 2/7 [00:12<00:31, 6.24s/it]\u001b[A\u001b[A\nTotal progress: 29%|██▊ | 2/7 [00:06<00:15, 3.08s/it]\u001b[A\nMultiDiffusion Sampling: 14%|█▍ | 2/14 [00:18<02:05, 10.44s/it]\u001b[A\n 43%|████▎ | 3/7 [00:18<00:24, 6.20s/it]\u001b[A\u001b[A\nTotal progress: 43%|████▎ | 3/7 [00:12<00:17, 4.36s/it]\u001b[A\nMultiDiffusion Sampling: 21%|██▏ | 3/14 [00:24<01:33, 8.49s/it]\u001b[A\n 57%|█████▋ | 4/7 [00:24<00:18, 6.18s/it]\u001b[A\u001b[A\nTotal progress: 57%|█████▋ | 4/7 [00:18<00:15, 5.03s/it]\u001b[A\nMultiDiffusion Sampling: 29%|██▊ | 4/14 [00:31<01:15, 7.57s/it]\u001b[A\n 71%|███████▏ | 5/7 [00:30<00:12, 6.17s/it]\u001b[A\u001b[A\nTotal progress: 71%|███████▏ | 5/7 [00:24<00:10, 5.42s/it]\u001b[A\nMultiDiffusion Sampling: 36%|███▌ | 5/14 [00:37<01:03, 7.06s/it]\u001b[A\n 86%|████████▌ | 6/7 [00:37<00:06, 6.17s/it]\u001b[A\u001b[A\nTotal progress: 86%|████████▌ | 6/7 [00:30<00:05, 5.67s/it]\u001b[A\nMultiDiffusion Sampling: 43%|████▎ | 6/14 [00:43<00:54, 6.75s/it]\u001b[A\n100%|██████████| 7/7 [00:43<00:00, 6.16s/it]\u001b[A\u001b[A\n100%|██████████| 7/7 [00:43<00:00, 6.19s/it]\nMultiDiffusion Sampling: 64%|██████▍ | 18/28 [03:19<01:50, 11.08s/it]\nTotal progress: 100%|██████████| 7/7 [00:36<00:00, 5.82s/it]\u001b[A[Tiled VAE]: input_size: torch.Size([1, 4, 364, 256]), tile_size: 192, padding: 11\n[Tiled VAE]: split to 2x2 = 4 tiles. Optimal tile size 128x192, original tile size 192x192\n[Tiled VAE]: Fast mode enabled, estimating group norm parameters on 135 x 192 image\n[Tiled VAE]: Executing Decoder Task Queue: 0%| | 0/492 [00:00<?, ?it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 25%|██▌ | 124/492 [00:00<00:01, 263.78it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 50%|█████ | 247/492 [00:00<00:00, 292.00it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 75%|███████▌ | 370/492 [00:01<00:00, 309.49it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 492/492 [00:01<00:00, 319.62it/s]\n[Tiled VAE]: Done in 2.516s, max VRAM alloc 7565.287 MB\nTotal progress: 100%|██████████| 7/7 [00:39<00:00, 5.82s/it]\u001b[A\nTotal progress: 100%|██████████| 7/7 [00:39<00:00, 5.70s/it]", "metrics": { "predict_time": 53.594466, "total_time": 248.74812 }, "output": [ "https://replicate.delivery/pbxt/0Ky96TRj3G7vJ9IjF8XEaXeQTUaWJkft0SJqKc4vQgtnvWgSA/1337-3d62d490-e278-11ee-9c8c-2a0a4d26c304.png" ], "started_at": "2024-03-15T03:00:03.125019Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/pojd5ydbfoozq3rtylgsidu6jm", "cancel": "https://api.replicate.com/v1/predictions/pojd5ydbfoozq3rtylgsidu6jm/cancel" }, "version": "96c34bbe9aae48023bb102b0386f62a88ecd05bcdac34e95ca10857af055e895" }Generated in[Tiled Diffusion] upscaling image with 4x-UltraSharp... [Tiled Diffusion] ControlNet found, support is enabled. 2024-03-15 03:00:05,196 - ControlNet - INFO - unit_separate = False, style_align = False 2024-03-15 03:00:05,196 - ControlNet - INFO - Loading model from cache: control_v11f1e_sd15_tile 2024-03-15 03:00:05,208 - ControlNet - INFO - Using preprocessor: tile_resample 2024-03-15 03:00:05,208 - ControlNet - INFO - preprocessor resolution = 2048 2024-03-15 03:00:05,309 - ControlNet - INFO - ControlNet Hooked - Time = 0.11690354347229004 MultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 9, Batch size: 5, Tile batches: 2 (ext: ContrlNet) [Tiled VAE]: the input size is tiny and unnecessary to tile. MultiDiffusion Sampling: 0%| | 0/14 [00:00<?, ?it/s] 0%| | 0/7 [00:00<?, ?it/s] MultiDiffusion Sampling: 7%|▋ | 1/14 [00:03<00:45, 3.46s/it] Total progress: 0%| | 0/7 [00:00<?, ?it/s] 14%|█▍ | 1/7 [00:06<00:38, 6.37s/it] 29%|██▊ | 2/7 [00:12<00:31, 6.24s/it] Total progress: 29%|██▊ | 2/7 [00:06<00:15, 3.08s/it] MultiDiffusion Sampling: 14%|█▍ | 2/14 [00:18<02:05, 10.44s/it] 43%|████▎ | 3/7 [00:18<00:24, 6.20s/it] Total progress: 43%|████▎ | 3/7 [00:12<00:17, 4.36s/it] MultiDiffusion Sampling: 21%|██▏ | 3/14 [00:24<01:33, 8.49s/it] 57%|█████▋ | 4/7 [00:24<00:18, 6.18s/it] Total progress: 57%|█████▋ | 4/7 [00:18<00:15, 5.03s/it] MultiDiffusion Sampling: 29%|██▊ | 4/14 [00:31<01:15, 7.57s/it] 71%|███████▏ | 5/7 [00:30<00:12, 6.17s/it] Total progress: 71%|███████▏ | 5/7 [00:24<00:10, 5.42s/it] MultiDiffusion Sampling: 36%|███▌ | 5/14 [00:37<01:03, 7.06s/it] 86%|████████▌ | 6/7 [00:37<00:06, 6.17s/it] Total progress: 86%|████████▌ | 6/7 [00:30<00:05, 5.67s/it] MultiDiffusion Sampling: 43%|████▎ | 6/14 [00:43<00:54, 6.75s/it] 100%|██████████| 7/7 [00:43<00:00, 6.16s/it] 100%|██████████| 7/7 [00:43<00:00, 6.19s/it] MultiDiffusion Sampling: 64%|██████▍ | 18/28 [03:19<01:50, 11.08s/it] Total progress: 100%|██████████| 7/7 [00:36<00:00, 5.82s/it][Tiled VAE]: input_size: torch.Size([1, 4, 364, 256]), tile_size: 192, padding: 11 [Tiled VAE]: split to 2x2 = 4 tiles. Optimal tile size 128x192, original tile size 192x192 [Tiled VAE]: Fast mode enabled, estimating group norm parameters on 135 x 192 image [Tiled VAE]: Executing Decoder Task Queue: 0%| | 0/492 [00:00<?, ?it/s] [Tiled VAE]: Executing Decoder Task Queue: 25%|██▌ | 124/492 [00:00<00:01, 263.78it/s] [Tiled VAE]: Executing Decoder Task Queue: 50%|█████ | 247/492 [00:00<00:00, 292.00it/s] [Tiled VAE]: Executing Decoder Task Queue: 75%|███████▌ | 370/492 [00:01<00:00, 309.49it/s] [Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 492/492 [00:01<00:00, 319.62it/s] [Tiled VAE]: Done in 2.516s, max VRAM alloc 7565.287 MB Total progress: 100%|██████████| 7/7 [00:39<00:00, 5.82s/it] Total progress: 100%|██████████| 7/7 [00:39<00:00, 5.70s/it]Prediction
philz1337x/clarity-upscaler:dfad4170ID5zkd3ulby644qrdmj3vnhxcz3yStatusSucceededSourceWebHardwareA40Total durationCreatedInput
- seed
- 1337
- image
- prompt
- masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>
- dynamic
- 6
- scheduler
- DPM++ 3M SDE Karras
- creativity
- 0.35
- resemblance
- 0.6
- scale_factor
- 4
- negative_prompt
- (worst quality, low quality, normal quality:2) JuggernautNegative-neg
- num_inference_steps
- 18
{ "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVPbpWlFJiM9y6eM9o13vHkQZRMTlsU9yOYS5ASZRzLZPjJ/img4originallow.jpg", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 }npm install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import and set up the clientimport Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", { input: { seed: 1337, image: "https://replicate.delivery/pbxt/KZVPbpWlFJiM9y6eM9o13vHkQZRMTlsU9yOYS5ASZRzLZPjJ/img4originallow.jpg", prompt: "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", dynamic: 6, scheduler: "DPM++ 3M SDE Karras", creativity: 0.35, resemblance: 0.6, scale_factor: 4, negative_prompt: "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", num_inference_steps: 18 } } ); console.log(output);To learn more, take a look at the guide on getting started with Node.js.
pip install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import the clientimport replicateRun philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", input={ "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVPbpWlFJiM9y6eM9o13vHkQZRMTlsU9yOYS5ASZRzLZPjJ/img4originallow.jpg", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 } ) print(output)To learn more, take a look at the guide on getting started with Python.
Set theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVPbpWlFJiM9y6eM9o13vHkQZRMTlsU9yOYS5ASZRzLZPjJ/img4originallow.jpg", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 } }' \ https://api.replicate.com/v1/predictionsTo learn more, take a look at Replicate’s HTTP API reference docs.
Output
Loading...
{ "completed_at": "2024-03-15T03:00:02.831458Z", "created_at": "2024-03-15T02:56:12.166062Z", "data_removed": false, "error": null, "id": "5zkd3ulby644qrdmj3vnhxcz3y", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVPbpWlFJiM9y6eM9o13vHkQZRMTlsU9yOYS5ASZRzLZPjJ/img4originallow.jpg", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 }, "logs": "[Tiled Diffusion] upscaling image with 4x-UltraSharp...\n[Tiled Diffusion] ControlNet found, support is enabled.\n2024-03-15 02:57:30,569 - ControlNet - \u001b[0;32mINFO\u001b[0m - unit_separate = False, style_align = False\n2024-03-15 02:57:30,569 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loading model from cache: control_v11f1e_sd15_tile\n2024-03-15 02:57:30,588 - ControlNet - \u001b[0;32mINFO\u001b[0m - Using preprocessor: tile_resample\n2024-03-15 02:57:30,588 - ControlNet - \u001b[0;32mINFO\u001b[0m - preprocessor resolution = 3584\n2024-03-15 02:57:30,834 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet Hooked - Time = 0.2695503234863281\nMultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 25, Batch size: 7, Tile batches: 4 (ext: ContrlNet)\nMultiDiffusion Sampling: 0%| | 0/28 [00:00<?, ?it/s]\nMultiDiffusion Sampling: 43%|████▎ | 6/14 [00:56<01:14, 9.34s/it]\n[Tiled VAE]: input_size: torch.Size([1, 3, 5376, 3584]), tile_size: 3072, padding: 32\n[Tiled VAE]: split to 2x2 = 4 tiles. Optimal tile size 1760x2656, original tile size 3072x3072\n[Tiled VAE]: Fast mode enabled, estimating group norm parameters on 2048 x 3072 image\n[Tiled VAE]: Executing Encoder Task Queue: 0%| | 0/364 [00:00<?, ?it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 5%|▌ | 19/364 [00:00<00:08, 41.10it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 7%|▋ | 24/364 [00:00<00:08, 41.95it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 10%|█ | 38/364 [00:01<00:09, 36.03it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 12%|█▏ | 42/364 [00:01<00:08, 36.22it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 16%|█▌ | 57/364 [00:01<00:08, 34.80it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 17%|█▋ | 61/364 [00:01<00:08, 35.43it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 21%|██ | 76/364 [00:02<00:08, 34.70it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 22%|██▏ | 80/364 [00:02<00:09, 29.20it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 23%|██▎ | 83/364 [00:02<00:11, 24.02it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 24%|██▎ | 86/364 [00:02<00:13, 20.36it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 25%|██▌ | 92/364 [00:03<00:11, 23.94it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 26%|██▌ | 95/364 [00:03<00:16, 15.90it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 27%|██▋ | 97/364 [00:03<00:16, 15.90it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 27%|██▋ | 99/364 [00:04<00:25, 10.39it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 28%|██▊ | 102/364 [00:04<00:21, 12.30it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 29%|██▊ | 104/364 [00:04<00:30, 8.55it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 29%|██▉ | 107/364 [00:04<00:24, 10.67it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 30%|██▉ | 109/364 [00:05<00:24, 10.27it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 31%|███ | 112/364 [00:05<00:24, 10.12it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 32%|███▏ | 115/364 [00:05<00:25, 9.79it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 32%|███▏ | 118/364 [00:06<00:25, 9.62it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 34%|███▍ | 124/364 [00:06<00:16, 14.50it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 35%|███▍ | 126/364 [00:06<00:16, 14.77it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 35%|███▌ | 128/364 [00:06<00:15, 14.99it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 36%|███▌ | 130/364 [00:06<00:15, 15.23it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 36%|███▋ | 132/364 [00:06<00:14, 15.94it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 37%|███▋ | 134/364 [00:06<00:15, 14.86it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 37%|███▋ | 136/364 [00:07<00:15, 15.09it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 38%|███▊ | 138/364 [00:07<00:14, 15.89it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 38%|███▊ | 140/364 [00:07<00:14, 15.29it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 39%|███▉ | 142/364 [00:07<00:14, 15.53it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 40%|███▉ | 144/364 [00:07<00:13, 15.89it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 41%|████ | 148/364 [00:07<00:11, 19.45it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 41%|████▏ | 151/364 [00:07<00:10, 19.86it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 42%|████▏ | 154/364 [00:07<00:10, 20.44it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 44%|████▍ | 161/364 [00:08<00:07, 26.98it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 45%|████▌ | 164/364 [00:08<00:07, 25.03it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 46%|████▌ | 167/364 [00:08<00:11, 17.63it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 47%|████▋ | 171/364 [00:08<00:10, 19.27it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 48%|████▊ | 174/364 [00:08<00:09, 19.57it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 49%|████▊ | 177/364 [00:09<00:09, 18.76it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 49%|████▉ | 180/364 [00:09<00:09, 18.89it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 50%|█████ | 183/364 [00:09<00:09, 18.55it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 51%|█████ | 186/364 [00:09<00:09, 18.14it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 53%|█████▎ | 192/364 [00:09<00:06, 26.39it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 54%|█████▍ | 198/364 [00:09<00:05, 29.70it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 55%|█████▌ | 202/364 [00:09<00:05, 31.38it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 57%|█████▋ | 206/364 [00:10<00:04, 33.10it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 58%|█████▊ | 210/364 [00:10<00:04, 31.66it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 60%|██████ | 219/364 [00:10<00:03, 43.00it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 63%|██████▎ | 230/364 [00:10<00:02, 55.47it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 66%|██████▌ | 239/364 [00:10<00:02, 60.65it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 68%|██████▊ | 248/364 [00:10<00:01, 63.78it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 71%|███████▏ | 260/364 [00:10<00:01, 75.83it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 74%|███████▎ | 268/364 [00:10<00:01, 69.40it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 76%|███████▌ | 276/364 [00:11<00:01, 68.62it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 79%|███████▊ | 286/364 [00:11<00:01, 71.63it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 82%|████████▏ | 297/364 [00:11<00:00, 81.09it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 84%|████████▍ | 306/364 [00:11<00:00, 74.39it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 86%|████████▋ | 314/364 [00:11<00:01, 46.67it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 88%|████████▊ | 321/364 [00:12<00:01, 26.80it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 90%|████████▉ | 326/364 [00:12<00:01, 24.80it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 92%|█████████▏| 334/364 [00:12<00:00, 30.81it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 95%|█████████▍| 345/364 [00:12<00:00, 41.52it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 99%|█████████▊| 359/364 [00:13<00:00, 57.88it/s]\u001b[A\n[Tiled VAE]: Executing Encoder Task Queue: 100%|██████████| 364/364 [00:13<00:00, 27.91it/s]\n[Tiled VAE]: Done in 13.930s, max VRAM alloc 26885.938 MB\n 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\nMultiDiffusion Sampling: 4%|▎ | 1/28 [00:17<07:51, 17.46s/it]\nMultiDiffusion Sampling: 7%|▋ | 2/28 [00:22<04:21, 10.05s/it]\nMultiDiffusion Sampling: 11%|█ | 3/28 [00:27<03:12, 7.68s/it]\nTotal progress: 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\n 14%|█▍ | 1/7 [00:17<01:46, 17.80s/it]\u001b[A\n 29%|██▊ | 2/7 [00:35<01:27, 17.58s/it]\u001b[A\nMultiDiffusion Sampling: 14%|█▍ | 4/28 [00:57<06:35, 16.47s/it]\nMultiDiffusion Sampling: 18%|█▊ | 5/28 [01:02<04:42, 12.30s/it]\nTotal progress: 29%|██▊ | 2/7 [00:17<00:43, 8.72s/it]\u001b[A\nMultiDiffusion Sampling: 21%|██▏ | 6/28 [01:05<03:26, 9.37s/it]\n 43%|████▎ | 3/7 [00:52<01:10, 17.52s/it]\u001b[A\nMultiDiffusion Sampling: 25%|██▌ | 7/28 [01:14<03:13, 9.20s/it]\nMultiDiffusion Sampling: 29%|██▊ | 8/28 [01:19<02:36, 7.83s/it]\nTotal progress: 43%|████▎ | 3/7 [00:34<00:49, 12.35s/it]\u001b[A\nMultiDiffusion Sampling: 32%|███▏ | 9/28 [01:23<02:04, 6.53s/it]\n 57%|█████▋ | 4/7 [01:10<00:52, 17.49s/it]\u001b[A\nMultiDiffusion Sampling: 36%|███▌ | 10/28 [01:32<02:10, 7.25s/it]\nMultiDiffusion Sampling: 39%|███▉ | 11/28 [01:36<01:51, 6.53s/it]\nTotal progress: 57%|█████▋ | 4/7 [00:52<00:42, 14.25s/it]\u001b[A\nMultiDiffusion Sampling: 43%|████▎ | 12/28 [01:40<01:30, 5.67s/it]\n 71%|███████▏ | 5/7 [01:27<00:34, 17.47s/it]\u001b[A\nMultiDiffusion Sampling: 46%|████▋ | 13/28 [01:49<01:39, 6.64s/it]\nMultiDiffusion Sampling: 50%|█████ | 14/28 [01:54<01:25, 6.11s/it]\nTotal progress: 71%|███████▏ | 5/7 [01:09<00:30, 15.36s/it]\u001b[A\nMultiDiffusion Sampling: 54%|█████▎ | 15/28 [01:58<01:09, 5.38s/it]\n 86%|████████▌ | 6/7 [01:45<00:17, 17.46s/it]\u001b[A\nMultiDiffusion Sampling: 57%|█████▋ | 16/28 [02:06<01:17, 6.43s/it]\nMultiDiffusion Sampling: 61%|██████ | 17/28 [02:11<01:05, 5.97s/it]\nTotal progress: 86%|████████▌ | 6/7 [01:27<00:16, 16.06s/it]\u001b[A\nMultiDiffusion Sampling: 64%|██████▍ | 18/28 [02:15<00:52, 5.28s/it]\n100%|██████████| 7/7 [02:02<00:00, 17.46s/it]\u001b[A\n100%|██████████| 7/7 [02:02<00:00, 17.49s/it]\nTotal progress: 100%|██████████| 7/7 [01:44<00:00, 16.50s/it]\u001b[A[Tiled VAE]: input_size: torch.Size([1, 4, 672, 448]), tile_size: 192, padding: 11\n[Tiled VAE]: split to 4x3 = 12 tiles. Optimal tile size 160x192, original tile size 192x192\n[Tiled VAE]: Fast mode enabled, estimating group norm parameters on 128 x 192 image\n[Tiled VAE]: Executing Decoder Task Queue: 0%| | 0/1476 [00:00<?, ?it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 8%|▊ | 124/1476 [00:00<00:06, 209.80it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 17%|█▋ | 247/1476 [00:01<00:05, 211.43it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 25%|██▌ | 370/1476 [00:01<00:04, 248.61it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 33%|███▎ | 493/1476 [00:02<00:04, 233.45it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 42%|████▏ | 616/1476 [00:02<00:03, 225.67it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 50%|█████ | 739/1476 [00:03<00:02, 249.81it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 58%|█████▊ | 862/1476 [00:03<00:02, 236.54it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 67%|██████▋ | 985/1476 [00:04<00:02, 228.78it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 75%|███████▌ | 1108/1476 [00:04<00:01, 250.21it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 83%|████████▎ | 1231/1476 [00:04<00:00, 296.13it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 92%|█████████▏| 1354/1476 [00:05<00:00, 338.84it/s]\u001b[A\u001b[A\n[Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 1476/1476 [00:05<00:00, 278.07it/s]\n[Tiled VAE]: Done in 6.394s, max VRAM alloc 9648.971 MB\nTotal progress: 100%|██████████| 7/7 [01:51<00:00, 16.50s/it]\u001b[A\nTotal progress: 100%|██████████| 7/7 [01:51<00:00, 15.98s/it]", "metrics": { "predict_time": 157.105595, "total_time": 230.665396 }, "output": [ "https://replicate.delivery/pbxt/2FDs9v66cH5SI1UEUCRMrPoQJMfb0jmPRxSlGfgPdk7wuWgSA/1337-1cd972b0-e278-11ee-9c8c-2a0a4d26c304.png" ], "started_at": "2024-03-15T02:57:25.725863Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/5zkd3ulby644qrdmj3vnhxcz3y", "cancel": "https://api.replicate.com/v1/predictions/5zkd3ulby644qrdmj3vnhxcz3y/cancel" }, "version": "96c34bbe9aae48023bb102b0386f62a88ecd05bcdac34e95ca10857af055e895" }Generated in[Tiled Diffusion] upscaling image with 4x-UltraSharp... [Tiled Diffusion] ControlNet found, support is enabled. 2024-03-15 02:57:30,569 - ControlNet - INFO - unit_separate = False, style_align = False 2024-03-15 02:57:30,569 - ControlNet - INFO - Loading model from cache: control_v11f1e_sd15_tile 2024-03-15 02:57:30,588 - ControlNet - INFO - Using preprocessor: tile_resample 2024-03-15 02:57:30,588 - ControlNet - INFO - preprocessor resolution = 3584 2024-03-15 02:57:30,834 - ControlNet - INFO - ControlNet Hooked - Time = 0.2695503234863281 MultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 25, Batch size: 7, Tile batches: 4 (ext: ContrlNet) MultiDiffusion Sampling: 0%| | 0/28 [00:00<?, ?it/s] MultiDiffusion Sampling: 43%|████▎ | 6/14 [00:56<01:14, 9.34s/it] [Tiled VAE]: input_size: torch.Size([1, 3, 5376, 3584]), tile_size: 3072, padding: 32 [Tiled VAE]: split to 2x2 = 4 tiles. Optimal tile size 1760x2656, original tile size 3072x3072 [Tiled VAE]: Fast mode enabled, estimating group norm parameters on 2048 x 3072 image [Tiled VAE]: Executing Encoder Task Queue: 0%| | 0/364 [00:00<?, ?it/s] [Tiled VAE]: Executing Encoder Task Queue: 5%|▌ | 19/364 [00:00<00:08, 41.10it/s] [Tiled VAE]: Executing Encoder Task Queue: 7%|▋ | 24/364 [00:00<00:08, 41.95it/s] [Tiled VAE]: Executing Encoder Task Queue: 10%|█ | 38/364 [00:01<00:09, 36.03it/s] [Tiled VAE]: Executing Encoder Task Queue: 12%|█▏ | 42/364 [00:01<00:08, 36.22it/s] [Tiled VAE]: Executing Encoder Task Queue: 16%|█▌ | 57/364 [00:01<00:08, 34.80it/s] [Tiled VAE]: Executing Encoder Task Queue: 17%|█▋ | 61/364 [00:01<00:08, 35.43it/s] [Tiled VAE]: Executing Encoder Task Queue: 21%|██ | 76/364 [00:02<00:08, 34.70it/s] [Tiled VAE]: Executing Encoder Task Queue: 22%|██▏ | 80/364 [00:02<00:09, 29.20it/s] [Tiled VAE]: Executing Encoder Task Queue: 23%|██▎ | 83/364 [00:02<00:11, 24.02it/s] [Tiled VAE]: Executing Encoder Task Queue: 24%|██▎ | 86/364 [00:02<00:13, 20.36it/s] [Tiled VAE]: Executing Encoder Task Queue: 25%|██▌ | 92/364 [00:03<00:11, 23.94it/s] [Tiled VAE]: Executing Encoder Task Queue: 26%|██▌ | 95/364 [00:03<00:16, 15.90it/s] [Tiled VAE]: Executing Encoder Task Queue: 27%|██▋ | 97/364 [00:03<00:16, 15.90it/s] [Tiled VAE]: Executing Encoder Task Queue: 27%|██▋ | 99/364 [00:04<00:25, 10.39it/s] [Tiled VAE]: Executing Encoder Task Queue: 28%|██▊ | 102/364 [00:04<00:21, 12.30it/s] [Tiled VAE]: Executing Encoder Task Queue: 29%|██▊ | 104/364 [00:04<00:30, 8.55it/s] [Tiled VAE]: Executing Encoder Task Queue: 29%|██▉ | 107/364 [00:04<00:24, 10.67it/s] [Tiled VAE]: Executing Encoder Task Queue: 30%|██▉ | 109/364 [00:05<00:24, 10.27it/s] [Tiled VAE]: Executing Encoder Task Queue: 31%|███ | 112/364 [00:05<00:24, 10.12it/s] [Tiled VAE]: Executing Encoder Task Queue: 32%|███▏ | 115/364 [00:05<00:25, 9.79it/s] [Tiled VAE]: Executing Encoder Task Queue: 32%|███▏ | 118/364 [00:06<00:25, 9.62it/s] [Tiled VAE]: Executing Encoder Task Queue: 34%|███▍ | 124/364 [00:06<00:16, 14.50it/s] [Tiled VAE]: Executing Encoder Task Queue: 35%|███▍ | 126/364 [00:06<00:16, 14.77it/s] [Tiled VAE]: Executing Encoder Task Queue: 35%|███▌ | 128/364 [00:06<00:15, 14.99it/s] [Tiled VAE]: Executing Encoder Task Queue: 36%|███▌ | 130/364 [00:06<00:15, 15.23it/s] [Tiled VAE]: Executing Encoder Task Queue: 36%|███▋ | 132/364 [00:06<00:14, 15.94it/s] [Tiled VAE]: Executing Encoder Task Queue: 37%|███▋ | 134/364 [00:06<00:15, 14.86it/s] [Tiled VAE]: Executing Encoder Task Queue: 37%|███▋ | 136/364 [00:07<00:15, 15.09it/s] [Tiled VAE]: Executing Encoder Task Queue: 38%|███▊ | 138/364 [00:07<00:14, 15.89it/s] [Tiled VAE]: Executing Encoder Task Queue: 38%|███▊ | 140/364 [00:07<00:14, 15.29it/s] [Tiled VAE]: Executing Encoder Task Queue: 39%|███▉ | 142/364 [00:07<00:14, 15.53it/s] [Tiled VAE]: Executing Encoder Task Queue: 40%|███▉ | 144/364 [00:07<00:13, 15.89it/s] [Tiled VAE]: Executing Encoder Task Queue: 41%|████ | 148/364 [00:07<00:11, 19.45it/s] [Tiled VAE]: Executing Encoder Task Queue: 41%|████▏ | 151/364 [00:07<00:10, 19.86it/s] [Tiled VAE]: Executing Encoder Task Queue: 42%|████▏ | 154/364 [00:07<00:10, 20.44it/s] [Tiled VAE]: Executing Encoder Task Queue: 44%|████▍ | 161/364 [00:08<00:07, 26.98it/s] [Tiled VAE]: Executing Encoder Task Queue: 45%|████▌ | 164/364 [00:08<00:07, 25.03it/s] [Tiled VAE]: Executing Encoder Task Queue: 46%|████▌ | 167/364 [00:08<00:11, 17.63it/s] [Tiled VAE]: Executing Encoder Task Queue: 47%|████▋ | 171/364 [00:08<00:10, 19.27it/s] [Tiled VAE]: Executing Encoder Task Queue: 48%|████▊ | 174/364 [00:08<00:09, 19.57it/s] [Tiled VAE]: Executing Encoder Task Queue: 49%|████▊ | 177/364 [00:09<00:09, 18.76it/s] [Tiled VAE]: Executing Encoder Task Queue: 49%|████▉ | 180/364 [00:09<00:09, 18.89it/s] [Tiled VAE]: Executing Encoder Task Queue: 50%|█████ | 183/364 [00:09<00:09, 18.55it/s] [Tiled VAE]: Executing Encoder Task Queue: 51%|█████ | 186/364 [00:09<00:09, 18.14it/s] [Tiled VAE]: Executing Encoder Task Queue: 53%|█████▎ | 192/364 [00:09<00:06, 26.39it/s] [Tiled VAE]: Executing Encoder Task Queue: 54%|█████▍ | 198/364 [00:09<00:05, 29.70it/s] [Tiled VAE]: Executing Encoder Task Queue: 55%|█████▌ | 202/364 [00:09<00:05, 31.38it/s] [Tiled VAE]: Executing Encoder Task Queue: 57%|█████▋ | 206/364 [00:10<00:04, 33.10it/s] [Tiled VAE]: Executing Encoder Task Queue: 58%|█████▊ | 210/364 [00:10<00:04, 31.66it/s] [Tiled VAE]: Executing Encoder Task Queue: 60%|██████ | 219/364 [00:10<00:03, 43.00it/s] [Tiled VAE]: Executing Encoder Task Queue: 63%|██████▎ | 230/364 [00:10<00:02, 55.47it/s] [Tiled VAE]: Executing Encoder Task Queue: 66%|██████▌ | 239/364 [00:10<00:02, 60.65it/s] [Tiled VAE]: Executing Encoder Task Queue: 68%|██████▊ | 248/364 [00:10<00:01, 63.78it/s] [Tiled VAE]: Executing Encoder Task Queue: 71%|███████▏ | 260/364 [00:10<00:01, 75.83it/s] [Tiled VAE]: Executing Encoder Task Queue: 74%|███████▎ | 268/364 [00:10<00:01, 69.40it/s] [Tiled VAE]: Executing Encoder Task Queue: 76%|███████▌ | 276/364 [00:11<00:01, 68.62it/s] [Tiled VAE]: Executing Encoder Task Queue: 79%|███████▊ | 286/364 [00:11<00:01, 71.63it/s] [Tiled VAE]: Executing Encoder Task Queue: 82%|████████▏ | 297/364 [00:11<00:00, 81.09it/s] [Tiled VAE]: Executing Encoder Task Queue: 84%|████████▍ | 306/364 [00:11<00:00, 74.39it/s] [Tiled VAE]: Executing Encoder Task Queue: 86%|████████▋ | 314/364 [00:11<00:01, 46.67it/s] [Tiled VAE]: Executing Encoder Task Queue: 88%|████████▊ | 321/364 [00:12<00:01, 26.80it/s] [Tiled VAE]: Executing Encoder Task Queue: 90%|████████▉ | 326/364 [00:12<00:01, 24.80it/s] [Tiled VAE]: Executing Encoder Task Queue: 92%|█████████▏| 334/364 [00:12<00:00, 30.81it/s] [Tiled VAE]: Executing Encoder Task Queue: 95%|█████████▍| 345/364 [00:12<00:00, 41.52it/s] [Tiled VAE]: Executing Encoder Task Queue: 99%|█████████▊| 359/364 [00:13<00:00, 57.88it/s] [Tiled VAE]: Executing Encoder Task Queue: 100%|██████████| 364/364 [00:13<00:00, 27.91it/s] [Tiled VAE]: Done in 13.930s, max VRAM alloc 26885.938 MB 0%| | 0/7 [00:00<?, ?it/s] MultiDiffusion Sampling: 4%|▎ | 1/28 [00:17<07:51, 17.46s/it] MultiDiffusion Sampling: 7%|▋ | 2/28 [00:22<04:21, 10.05s/it] MultiDiffusion Sampling: 11%|█ | 3/28 [00:27<03:12, 7.68s/it] Total progress: 0%| | 0/7 [00:00<?, ?it/s] 14%|█▍ | 1/7 [00:17<01:46, 17.80s/it] 29%|██▊ | 2/7 [00:35<01:27, 17.58s/it] MultiDiffusion Sampling: 14%|█▍ | 4/28 [00:57<06:35, 16.47s/it] MultiDiffusion Sampling: 18%|█▊ | 5/28 [01:02<04:42, 12.30s/it] Total progress: 29%|██▊ | 2/7 [00:17<00:43, 8.72s/it] MultiDiffusion Sampling: 21%|██▏ | 6/28 [01:05<03:26, 9.37s/it] 43%|████▎ | 3/7 [00:52<01:10, 17.52s/it] MultiDiffusion Sampling: 25%|██▌ | 7/28 [01:14<03:13, 9.20s/it] MultiDiffusion Sampling: 29%|██▊ | 8/28 [01:19<02:36, 7.83s/it] Total progress: 43%|████▎ | 3/7 [00:34<00:49, 12.35s/it] MultiDiffusion Sampling: 32%|███▏ | 9/28 [01:23<02:04, 6.53s/it] 57%|█████▋ | 4/7 [01:10<00:52, 17.49s/it] MultiDiffusion Sampling: 36%|███▌ | 10/28 [01:32<02:10, 7.25s/it] MultiDiffusion Sampling: 39%|███▉ | 11/28 [01:36<01:51, 6.53s/it] Total progress: 57%|█████▋ | 4/7 [00:52<00:42, 14.25s/it] MultiDiffusion Sampling: 43%|████▎ | 12/28 [01:40<01:30, 5.67s/it] 71%|███████▏ | 5/7 [01:27<00:34, 17.47s/it] MultiDiffusion Sampling: 46%|████▋ | 13/28 [01:49<01:39, 6.64s/it] MultiDiffusion Sampling: 50%|█████ | 14/28 [01:54<01:25, 6.11s/it] Total progress: 71%|███████▏ | 5/7 [01:09<00:30, 15.36s/it] MultiDiffusion Sampling: 54%|█████▎ | 15/28 [01:58<01:09, 5.38s/it] 86%|████████▌ | 6/7 [01:45<00:17, 17.46s/it] MultiDiffusion Sampling: 57%|█████▋ | 16/28 [02:06<01:17, 6.43s/it] MultiDiffusion Sampling: 61%|██████ | 17/28 [02:11<01:05, 5.97s/it] Total progress: 86%|████████▌ | 6/7 [01:27<00:16, 16.06s/it] MultiDiffusion Sampling: 64%|██████▍ | 18/28 [02:15<00:52, 5.28s/it] 100%|██████████| 7/7 [02:02<00:00, 17.46s/it] 100%|██████████| 7/7 [02:02<00:00, 17.49s/it] Total progress: 100%|██████████| 7/7 [01:44<00:00, 16.50s/it][Tiled VAE]: input_size: torch.Size([1, 4, 672, 448]), tile_size: 192, padding: 11 [Tiled VAE]: split to 4x3 = 12 tiles. Optimal tile size 160x192, original tile size 192x192 [Tiled VAE]: Fast mode enabled, estimating group norm parameters on 128 x 192 image [Tiled VAE]: Executing Decoder Task Queue: 0%| | 0/1476 [00:00<?, ?it/s] [Tiled VAE]: Executing Decoder Task Queue: 8%|▊ | 124/1476 [00:00<00:06, 209.80it/s] [Tiled VAE]: Executing Decoder Task Queue: 17%|█▋ | 247/1476 [00:01<00:05, 211.43it/s] [Tiled VAE]: Executing Decoder Task Queue: 25%|██▌ | 370/1476 [00:01<00:04, 248.61it/s] [Tiled VAE]: Executing Decoder Task Queue: 33%|███▎ | 493/1476 [00:02<00:04, 233.45it/s] [Tiled VAE]: Executing Decoder Task Queue: 42%|████▏ | 616/1476 [00:02<00:03, 225.67it/s] [Tiled VAE]: Executing Decoder Task Queue: 50%|█████ | 739/1476 [00:03<00:02, 249.81it/s] [Tiled VAE]: Executing Decoder Task Queue: 58%|█████▊ | 862/1476 [00:03<00:02, 236.54it/s] [Tiled VAE]: Executing Decoder Task Queue: 67%|██████▋ | 985/1476 [00:04<00:02, 228.78it/s] [Tiled VAE]: Executing Decoder Task Queue: 75%|███████▌ | 1108/1476 [00:04<00:01, 250.21it/s] [Tiled VAE]: Executing Decoder Task Queue: 83%|████████▎ | 1231/1476 [00:04<00:00, 296.13it/s] [Tiled VAE]: Executing Decoder Task Queue: 92%|█████████▏| 1354/1476 [00:05<00:00, 338.84it/s] [Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 1476/1476 [00:05<00:00, 278.07it/s] [Tiled VAE]: Done in 6.394s, max VRAM alloc 9648.971 MB Total progress: 100%|██████████| 7/7 [01:51<00:00, 16.50s/it] Total progress: 100%|██████████| 7/7 [01:51<00:00, 15.98s/it]Prediction
philz1337x/clarity-upscaler:dfad4170IDqqh7vsdb4nmsd7r6icmjcqe5heStatusSucceededSourceWebHardwareA40Total durationCreatedInput
- seed
- 1337
- image
- prompt
- masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>
- dynamic
- 6
- scheduler
- DPM++ 3M SDE Karras
- creativity
- 0.35
- resemblance
- 0.6
- scale_factor
- 4
- negative_prompt
- (worst quality, low quality, normal quality:2) JuggernautNegative-neg
- num_inference_steps
- 18
{ "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVNCfqUqyMBvHD9qiKli5gSrmuYftalyLss6LhI2ctwOW0H/3_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 }npm install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import and set up the clientimport Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", { input: { seed: 1337, image: "https://replicate.delivery/pbxt/KZVNCfqUqyMBvHD9qiKli5gSrmuYftalyLss6LhI2ctwOW0H/3_before.png", prompt: "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", dynamic: 6, scheduler: "DPM++ 3M SDE Karras", creativity: 0.35, resemblance: 0.6, scale_factor: 4, negative_prompt: "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", num_inference_steps: 18 } } ); console.log(output);To learn more, take a look at the guide on getting started with Node.js.
pip install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import the clientimport replicateRun philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", input={ "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVNCfqUqyMBvHD9qiKli5gSrmuYftalyLss6LhI2ctwOW0H/3_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 } ) print(output)To learn more, take a look at the guide on getting started with Python.
Set theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVNCfqUqyMBvHD9qiKli5gSrmuYftalyLss6LhI2ctwOW0H/3_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 } }' \ https://api.replicate.com/v1/predictionsTo learn more, take a look at Replicate’s HTTP API reference docs.
Output
Loading...
{ "completed_at": "2024-03-15T03:07:26.247084Z", "created_at": "2024-03-15T03:04:55.946651Z", "data_removed": false, "error": null, "id": "qqh7vsdb4nmsd7r6icmjcqe5he", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KZVNCfqUqyMBvHD9qiKli5gSrmuYftalyLss6LhI2ctwOW0H/3_before.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "resemblance": 0.6, "scale_factor": 4, "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18 }, "logs": "Creating model from config: /src/configs/v1-inference.yaml\n2024-03-15 03:06:27,506 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet UI callback registered.\nfatal: not a git repository (or any of the parent directories): .git\nfatal: not a git repository (or any of the parent directories): .git\nCouldn't find VAE named None; using None instead\nApplying attention optimization: Doggettx... done.\nModel loaded in 3.4s (load weights from disk: 0.1s, create model: 2.4s, apply weights to model: 0.6s).\nLoading VAE weights specified in settings: /src/models/VAE/vae-ft-mse-840000-ema-pruned.safetensors\nApplying attention optimization: Doggettx... done.\nVAE weights loaded.\n[Tiled Diffusion] upscaling image with 4x-UltraSharp...\n[Tiled Diffusion] ControlNet found, support is enabled.\n2024-03-15 03:06:32,954 - ControlNet - \u001b[0;32mINFO\u001b[0m - unit_separate = False, style_align = False\n2024-03-15 03:06:33,218 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loading model: control_v11f1e_sd15_tile [a371b31b]\n2024-03-15 03:06:33,546 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loaded state_dict from [/src/extensions/sd-webui-controlnet/models/control_v11f1e_sd15_tile.pth]\n2024-03-15 03:06:33,546 - ControlNet - \u001b[0;32mINFO\u001b[0m - controlnet_default_config\n2024-03-15 03:06:35,821 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet model control_v11f1e_sd15_tile [a371b31b](ControlModelType.ControlNet) loaded.\n2024-03-15 03:06:35,836 - ControlNet - \u001b[0;32mINFO\u001b[0m - Using preprocessor: tile_resample\n2024-03-15 03:06:35,836 - ControlNet - \u001b[0;32mINFO\u001b[0m - preprocessor resolution = 2048\n2024-03-15 03:06:35,941 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet Hooked - Time = 2.991502046585083\nMultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 8, Batch size: 8, Tile batches: 1 (ext: ContrlNet)\n[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling: : 0it [00:00, ?it/s]\n 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\nMultiDiffusion Sampling: : 0it [00:08, ?it/s]\nTotal progress: 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\n 14%|█▍ | 1/7 [00:06<00:39, 6.56s/it]\u001b[A\nTotal progress: 29%|██▊ | 2/7 [00:05<00:14, 2.82s/it]\u001b[A\n 29%|██▊ | 2/7 [00:12<00:30, 6.02s/it]\u001b[A\nTotal progress: 43%|████▎ | 3/7 [00:11<00:15, 4.00s/it]\u001b[A\n 43%|████▎ | 3/7 [00:17<00:23, 5.85s/it]\u001b[A\nTotal progress: 57%|█████▋ | 4/7 [00:16<00:13, 4.61s/it]\u001b[A\n 57%|█████▋ | 4/7 [00:23<00:17, 5.77s/it]\u001b[A\nTotal progress: 71%|███████▏ | 5/7 [00:22<00:09, 4.97s/it]\u001b[A\n 71%|███████▏ | 5/7 [00:29<00:11, 5.73s/it]\u001b[A\nTotal progress: 86%|████████▌ | 6/7 [00:28<00:05, 5.20s/it]\u001b[A\n 86%|████████▌ | 6/7 [00:34<00:05, 5.70s/it]\u001b[A\n100%|██████████| 7/7 [00:40<00:00, 5.68s/it]\u001b[A\n100%|██████████| 7/7 [00:40<00:00, 5.78s/it]\nTotal progress: 100%|██████████| 7/7 [00:33<00:00, 5.34s/it]\u001b[A[Tiled VAE]: input_size: torch.Size([1, 4, 256, 384]), tile_size: 192, padding: 11\n[Tiled VAE]: split to 2x2 = 4 tiles. Optimal tile size 192x128, original tile size 192x192\n[Tiled VAE]: Fast mode enabled, estimating group norm parameters on 192 x 128 image\n[Tiled VAE]: Executing Decoder Task Queue: 0%| | 0/492 [00:00<?, ?it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 25%|██▌ | 124/492 [00:00<00:01, 260.18it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 50%|█████ | 247/492 [00:00<00:00, 279.59it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 75%|███████▌ | 370/492 [00:01<00:00, 293.63it/s]\n[Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 492/492 [00:01<00:00, 300.73it/s]\n[Tiled VAE]: Done in 2.630s, max VRAM alloc 7654.307 MB\nTotal progress: 100%|██████████| 7/7 [00:37<00:00, 5.34s/it]\u001b[A\nTotal progress: 100%|██████████| 7/7 [00:37<00:00, 5.29s/it]", "metrics": { "predict_time": 59.288305, "total_time": 150.300433 }, "output": [ "https://replicate.delivery/pbxt/eff95gA2xvDTeTFGK4cYhwvMacBeona3xiz95C9zdd1it2CUC/1337-25402d3a-e279-11ee-9962-0631823f4226.png" ], "started_at": "2024-03-15T03:06:26.958779Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/qqh7vsdb4nmsd7r6icmjcqe5he", "cancel": "https://api.replicate.com/v1/predictions/qqh7vsdb4nmsd7r6icmjcqe5he/cancel" }, "version": "96c34bbe9aae48023bb102b0386f62a88ecd05bcdac34e95ca10857af055e895" }Generated inCreating model from config: /src/configs/v1-inference.yaml 2024-03-15 03:06:27,506 - ControlNet - INFO - ControlNet UI callback registered. fatal: not a git repository (or any of the parent directories): .git fatal: not a git repository (or any of the parent directories): .git Couldn't find VAE named None; using None instead Applying attention optimization: Doggettx... done. Model loaded in 3.4s (load weights from disk: 0.1s, create model: 2.4s, apply weights to model: 0.6s). Loading VAE weights specified in settings: /src/models/VAE/vae-ft-mse-840000-ema-pruned.safetensors Applying attention optimization: Doggettx... done. VAE weights loaded. [Tiled Diffusion] upscaling image with 4x-UltraSharp... [Tiled Diffusion] ControlNet found, support is enabled. 2024-03-15 03:06:32,954 - ControlNet - INFO - unit_separate = False, style_align = False 2024-03-15 03:06:33,218 - ControlNet - INFO - Loading model: control_v11f1e_sd15_tile [a371b31b] 2024-03-15 03:06:33,546 - ControlNet - INFO - Loaded state_dict from [/src/extensions/sd-webui-controlnet/models/control_v11f1e_sd15_tile.pth] 2024-03-15 03:06:33,546 - ControlNet - INFO - controlnet_default_config 2024-03-15 03:06:35,821 - ControlNet - INFO - ControlNet model control_v11f1e_sd15_tile [a371b31b](ControlModelType.ControlNet) loaded. 2024-03-15 03:06:35,836 - ControlNet - INFO - Using preprocessor: tile_resample 2024-03-15 03:06:35,836 - ControlNet - INFO - preprocessor resolution = 2048 2024-03-15 03:06:35,941 - ControlNet - INFO - ControlNet Hooked - Time = 2.991502046585083 MultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 8, Batch size: 8, Tile batches: 1 (ext: ContrlNet) [Tiled VAE]: the input size is tiny and unnecessary to tile. MultiDiffusion Sampling: : 0it [00:00, ?it/s] 0%| | 0/7 [00:00<?, ?it/s] MultiDiffusion Sampling: : 0it [00:08, ?it/s] Total progress: 0%| | 0/7 [00:00<?, ?it/s] 14%|█▍ | 1/7 [00:06<00:39, 6.56s/it] Total progress: 29%|██▊ | 2/7 [00:05<00:14, 2.82s/it] 29%|██▊ | 2/7 [00:12<00:30, 6.02s/it] Total progress: 43%|████▎ | 3/7 [00:11<00:15, 4.00s/it] 43%|████▎ | 3/7 [00:17<00:23, 5.85s/it] Total progress: 57%|█████▋ | 4/7 [00:16<00:13, 4.61s/it] 57%|█████▋ | 4/7 [00:23<00:17, 5.77s/it] Total progress: 71%|███████▏ | 5/7 [00:22<00:09, 4.97s/it] 71%|███████▏ | 5/7 [00:29<00:11, 5.73s/it] Total progress: 86%|████████▌ | 6/7 [00:28<00:05, 5.20s/it] 86%|████████▌ | 6/7 [00:34<00:05, 5.70s/it] 100%|██████████| 7/7 [00:40<00:00, 5.68s/it] 100%|██████████| 7/7 [00:40<00:00, 5.78s/it] Total progress: 100%|██████████| 7/7 [00:33<00:00, 5.34s/it][Tiled VAE]: input_size: torch.Size([1, 4, 256, 384]), tile_size: 192, padding: 11 [Tiled VAE]: split to 2x2 = 4 tiles. Optimal tile size 192x128, original tile size 192x192 [Tiled VAE]: Fast mode enabled, estimating group norm parameters on 192 x 128 image [Tiled VAE]: Executing Decoder Task Queue: 0%| | 0/492 [00:00<?, ?it/s] [Tiled VAE]: Executing Decoder Task Queue: 25%|██▌ | 124/492 [00:00<00:01, 260.18it/s] [Tiled VAE]: Executing Decoder Task Queue: 50%|█████ | 247/492 [00:00<00:00, 279.59it/s] [Tiled VAE]: Executing Decoder Task Queue: 75%|███████▌ | 370/492 [00:01<00:00, 293.63it/s] [Tiled VAE]: Executing Decoder Task Queue: 100%|██████████| 492/492 [00:01<00:00, 300.73it/s] [Tiled VAE]: Done in 2.630s, max VRAM alloc 7654.307 MB Total progress: 100%|██████████| 7/7 [00:37<00:00, 5.34s/it] Total progress: 100%|██████████| 7/7 [00:37<00:00, 5.29s/it]Prediction
philz1337x/clarity-upscaler:dfad4170Input
- seed
- 1337
- image
- prompt
- masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>
- dynamic
- 6
- sd_model
- juggernaut_reborn.safetensors [338b85bc4f]
- scheduler
- DPM++ 3M SDE Karras
- creativity
- 0.35
- lora_links
- downscaling
- resemblance
- 0.6
- scale_factor
- 2
- tiling_width
- 112
- tiling_height
- 144
- custom_sd_model
- negative_prompt
- (worst quality, low quality, normal quality:2) JuggernautNegative-neg
- num_inference_steps
- 18
- downscaling_resolution
- 768
{ "seed": 1337, "image": "https://replicate.delivery/pbxt/KiDB5iqtTcxiTI17WASotG1Ei0TNJCztdU6J02pnMYAd8B1X/13_before-4.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "lora_links": "", "downscaling": false, "resemblance": 0.6, "scale_factor": 2, "tiling_width": 112, "tiling_height": 144, "custom_sd_model": "", "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 768 }npm install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import and set up the clientimport Replicate from "replicate"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", { input: { seed: 1337, image: "https://replicate.delivery/pbxt/KiDB5iqtTcxiTI17WASotG1Ei0TNJCztdU6J02pnMYAd8B1X/13_before-4.png", prompt: "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", dynamic: 6, sd_model: "juggernaut_reborn.safetensors [338b85bc4f]", scheduler: "DPM++ 3M SDE Karras", creativity: 0.35, lora_links: "", downscaling: false, resemblance: 0.6, scale_factor: 2, tiling_width: 112, tiling_height: 144, custom_sd_model: "", negative_prompt: "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", num_inference_steps: 18, downscaling_resolution: 768 } } ); console.log(output);To learn more, take a look at the guide on getting started with Node.js.
pip install replicateSet theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Import the clientimport replicateRun philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "philz1337x/clarity-upscaler:dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", input={ "seed": 1337, "image": "https://replicate.delivery/pbxt/KiDB5iqtTcxiTI17WASotG1Ei0TNJCztdU6J02pnMYAd8B1X/13_before-4.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "lora_links": "", "downscaling": False, "resemblance": 0.6, "scale_factor": 2, "tiling_width": 112, "tiling_height": 144, "custom_sd_model": "", "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 768 } ) print(output)To learn more, take a look at the guide on getting started with Python.
Set theREPLICATE_API_TOKENenvironment variableexport REPLICATE_API_TOKEN=<paste-your-token-here>Find your API token in your account settings.
Run philz1337x/clarity-upscaler using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
curl -s -X POST \ -H "Authorization: Bearer $REPLICATE_API_TOKEN" \ -H "Content-Type: application/json" \ -H "Prefer: wait" \ -d $'{ "version": "dfad41707589d68ecdccd1dfa600d55a208f9310748e44bfe35b4a6291453d5e", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KiDB5iqtTcxiTI17WASotG1Ei0TNJCztdU6J02pnMYAd8B1X/13_before-4.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "lora_links": "", "downscaling": false, "resemblance": 0.6, "scale_factor": 2, "tiling_width": 112, "tiling_height": 144, "custom_sd_model": "", "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 768 } }' \ https://api.replicate.com/v1/predictionsTo learn more, take a look at Replicate’s HTTP API reference docs.
Output
Loading...
{ "completed_at": "2024-04-08T15:34:51.385820Z", "created_at": "2024-04-08T15:34:38.558000Z", "data_removed": false, "error": null, "id": "snbeynewvsrgp0ceqsc8gjk6sm", "input": { "seed": 1337, "image": "https://replicate.delivery/pbxt/KiDB5iqtTcxiTI17WASotG1Ei0TNJCztdU6J02pnMYAd8B1X/13_before-4.png", "prompt": "masterpiece, best quality, highres, <lora:more_details:0.5> <lora:SDXLrender_v2.0:1>", "dynamic": 6, "sd_model": "juggernaut_reborn.safetensors [338b85bc4f]", "scheduler": "DPM++ 3M SDE Karras", "creativity": 0.35, "lora_links": "", "downscaling": false, "resemblance": 0.6, "scale_factor": 2, "tiling_width": 112, "tiling_height": 144, "custom_sd_model": "", "negative_prompt": "(worst quality, low quality, normal quality:2) JuggernautNegative-neg", "num_inference_steps": 18, "downscaling_resolution": 768 }, "logs": "Running prediction\n[Tiled Diffusion] upscaling image with 4x-UltraSharp...\n[Tiled Diffusion] ControlNet found, support is enabled.\n2024-04-08 15:34:42,215 - ControlNet - \u001b[0;32mINFO\u001b[0m - unit_separate = False, style_align = False\n2024-04-08 15:34:42,215 - ControlNet - \u001b[0;32mINFO\u001b[0m - Loading model from cache: control_v11f1e_sd15_tile\n2024-04-08 15:34:42,234 - ControlNet - \u001b[0;32mINFO\u001b[0m - Using preprocessor: tile_resample\n2024-04-08 15:34:42,234 - ControlNet - \u001b[0;32mINFO\u001b[0m - preprocessor resolution = 1536\n2024-04-08 15:34:42,310 - ControlNet - \u001b[0;32mINFO\u001b[0m - ControlNet Hooked - Time = 0.0993812084197998\nMultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 4, Batch size: 4, Tile batches: 1 (ext: ContrlNet)\n[Tiled VAE]: the input size is tiny and unnecessary to tile.\nMultiDiffusion Sampling: 0%| | 0/10 [00:00<?, ?it/s]\n 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\nTotal progress: 0%| | 0/7 [00:00<?, ?it/s]\u001b[A\n 14%|█▍ | 1/7 [00:00<00:05, 1.10it/s]\u001b[A\nTotal progress: 29%|██▊ | 2/7 [00:00<00:02, 2.22it/s]\u001b[A\n 29%|██▊ | 2/7 [00:01<00:04, 1.11it/s]\u001b[A\nTotal progress: 43%|████▎ | 3/7 [00:01<00:02, 1.57it/s]\u001b[A\n 43%|████▎ | 3/7 [00:02<00:03, 1.11it/s]\u001b[A\nTotal progress: 57%|█████▋ | 4/7 [00:02<00:02, 1.36it/s]\u001b[A\n 57%|█████▋ | 4/7 [00:03<00:02, 1.11it/s]\u001b[A\nTotal progress: 71%|███████▏ | 5/7 [00:03<00:01, 1.26it/s]\u001b[A\n 71%|███████▏ | 5/7 [00:04<00:01, 1.11it/s]\u001b[A\nTotal progress: 86%|████████▌ | 6/7 [00:04<00:00, 1.21it/s]\u001b[A\n 86%|████████▌ | 6/7 [00:05<00:00, 1.11it/s]\u001b[A\n100%|██████████| 7/7 [00:06<00:00, 1.11it/s]\u001b[A\n100%|██████████| 7/7 [00:06<00:00, 1.11it/s]\nTotal progress: 100%|██████████| 7/7 [00:05<00:00, 1.18it/s]\u001b[A[Tiled VAE]: the input size is tiny and unnecessary to tile.\nTotal progress: 100%|██████████| 7/7 [00:06<00:00, 1.18it/s]\u001b[A\nTotal progress: 100%|██████████| 7/7 [00:06<00:00, 1.11it/s]\nPrediction took 11.742665767669678 seconds", "metrics": { "predict_time": 12.806754, "total_time": 12.82782 }, "output": [ "https://replicate.delivery/pbxt/00paK5p2fY36XykEcte0skfqnSl3akRGTm4TC2l2wiU0E4QlA/1337-89fd2a6e-f5bd-11ee-87b9-bec9d6ff70a9.png" ], "started_at": "2024-04-08T15:34:38.579066Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/snbeynewvsrgp0ceqsc8gjk6sm", "cancel": "https://api.replicate.com/v1/predictions/snbeynewvsrgp0ceqsc8gjk6sm/cancel" }, "version": "3bb9d3412f14261c2f8cfa15a56dd7b33f44af54777c517b860a3f678a5d7f3b" }Generated inRunning prediction [Tiled Diffusion] upscaling image with 4x-UltraSharp... [Tiled Diffusion] ControlNet found, support is enabled. 2024-04-08 15:34:42,215 - ControlNet - INFO - unit_separate = False, style_align = False 2024-04-08 15:34:42,215 - ControlNet - INFO - Loading model from cache: control_v11f1e_sd15_tile 2024-04-08 15:34:42,234 - ControlNet - INFO - Using preprocessor: tile_resample 2024-04-08 15:34:42,234 - ControlNet - INFO - preprocessor resolution = 1536 2024-04-08 15:34:42,310 - ControlNet - INFO - ControlNet Hooked - Time = 0.0993812084197998 MultiDiffusion hooked into 'DPM++ 3M SDE Karras' sampler, Tile size: 144x112, Tile count: 4, Batch size: 4, Tile batches: 1 (ext: ContrlNet) [Tiled VAE]: the input size is tiny and unnecessary to tile. MultiDiffusion Sampling: 0%| | 0/10 [00:00<?, ?it/s] 0%| | 0/7 [00:00<?, ?it/s] Total progress: 0%| | 0/7 [00:00<?, ?it/s] 14%|█▍ | 1/7 [00:00<00:05, 1.10it/s] Total progress: 29%|██▊ | 2/7 [00:00<00:02, 2.22it/s] 29%|██▊ | 2/7 [00:01<00:04, 1.11it/s] Total progress: 43%|████▎ | 3/7 [00:01<00:02, 1.57it/s] 43%|████▎ | 3/7 [00:02<00:03, 1.11it/s] Total progress: 57%|█████▋ | 4/7 [00:02<00:02, 1.36it/s] 57%|█████▋ | 4/7 [00:03<00:02, 1.11it/s] Total progress: 71%|███████▏ | 5/7 [00:03<00:01, 1.26it/s] 71%|███████▏ | 5/7 [00:04<00:01, 1.11it/s] Total progress: 86%|████████▌ | 6/7 [00:04<00:00, 1.21it/s] 86%|████████▌ | 6/7 [00:05<00:00, 1.11it/s] 100%|██████████| 7/7 [00:06<00:00, 1.11it/s] 100%|██████████| 7/7 [00:06<00:00, 1.11it/s] Total progress: 100%|██████████| 7/7 [00:05<00:00, 1.18it/s][Tiled VAE]: the input size is tiny and unnecessary to tile. Total progress: 100%|██████████| 7/7 [00:06<00:00, 1.18it/s] Total progress: 100%|██████████| 7/7 [00:06<00:00, 1.11it/s] Prediction took 11.742665767669678 seconds
Want to make some of these yourself?
Run this model