lucataco/sdxl-niji-se

SDXL_Niji_Special Edition

Shiba stable diffusion model

bigcode/tiny_starcoder_py

lmsys/vicuna-13b-v1.3

lmsys/vicuna-7b-v1.3

Salesforce/codegen2-1B

Salesforce/xgen-7b-8k-base

Realistic Vision V3.0 with VAE

Realistic Vision V4.0

CLIP Interrogator (for faster inference)
A working wsrglow model

RiversHaveWings Stable Diffusion Upscaler

Real-ESRGAN with optional face correction and adjustable upscale (for larger images)
Animate Your Personalized Text-to-Image Diffusion Models

Segments an audio recording based on who is speaking (on A100)

Meta's Llama 2 7b Chat - GPTQ

Meta's Llama 2 13b Chat - GPTQ

Stability AI's FreeWilly2

Implementation of Realistic Vision v5.1 with VAE

Practical face restoration algorithm for *old photos* or *AI-generated faces* (for larger images)

SDXL v1.0 - A text-to-image generative AI model that creates beautiful images
Prediction
lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449cIDq77swwlbbeyhicvbp32cxf24dmStatusSucceededSourceWebHardwareA100 (40GB)Total durationCreatedInput
- seed
- 201961
- width
- 1024
- height
- 1024
- prompt
- illustration, anime, Monkey D. Luffy, One Piece, vintage, port city, poster, in the style of [james jean], deathcore, [john pitre], detailed atmospheric portraits, oil painting, anime, biopunk, horror art, dark and intricate, unique illustrations, precisionist art, airbrush, mystical landscape, intricate psychedelic, dark azure and red, witchcore
- scheduler
- K_EULER
- num_outputs
- 1
- guidance_scale
- 7
- apply_watermark
- negative_prompt
- (worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31
- num_inference_steps
- 24
{ "seed": 201961, "width": 1024, "height": 1024, "prompt": "illustration, anime, Monkey D. Luffy, One Piece, vintage, port city, poster, in the style of [james jean], deathcore, [john pitre], detailed atmospheric portraits, oil painting, anime, biopunk, horror art, dark and intricate, unique illustrations, precisionist art, airbrush, mystical landscape, intricate psychedelic, dark azure and red, witchcore", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; import fs from "node:fs"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", { input: { seed: 201961, width: 1024, height: 1024, prompt: "illustration, anime, Monkey D. Luffy, One Piece, vintage, port city, poster, in the style of [james jean], deathcore, [john pitre], detailed atmospheric portraits, oil painting, anime, biopunk, horror art, dark and intricate, unique illustrations, precisionist art, airbrush, mystical landscape, intricate psychedelic, dark azure and red, witchcore", scheduler: "K_EULER", num_outputs: 1, guidance_scale: 7, apply_watermark: true, negative_prompt: "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", num_inference_steps: 24 } } ); // To access the file URL: console.log(output[0].url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output[0]);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", input={ "seed": 201961, "width": 1024, "height": 1024, "prompt": "illustration, anime, Monkey D. Luffy, One Piece, vintage, port city, poster, in the style of [james jean], deathcore, [john pitre], detailed atmospheric portraits, oil painting, anime, biopunk, horror art, dark and intricate, unique illustrations, precisionist art, airbrush, mystical landscape, intricate psychedelic, dark azure and red, witchcore", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": True, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 } ) # To access the file URL: print(output[0].url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output[0].read())
To learn more, take a look at the guide on getting started with Python.
Run lucataco/sdxl-niji-se 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": "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", "input": { "seed": 201961, "width": 1024, "height": 1024, "prompt": "illustration, anime, Monkey D. Luffy, One Piece, vintage, port city, poster, in the style of [james jean], deathcore, [john pitre], detailed atmospheric portraits, oil painting, anime, biopunk, horror art, dark and intricate, unique illustrations, precisionist art, airbrush, mystical landscape, intricate psychedelic, dark azure and red, witchcore", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you don’t have Homebrew, there are other installation options available.
Run this to download the model and run it in your local environment:
cog predict r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c \ -i 'seed=201961' \ -i 'width=1024' \ -i 'height=1024' \ -i 'prompt="illustration, anime, Monkey D. Luffy, One Piece, vintage, port city, poster, in the style of [james jean], deathcore, [john pitre], detailed atmospheric portraits, oil painting, anime, biopunk, horror art, dark and intricate, unique illustrations, precisionist art, airbrush, mystical landscape, intricate psychedelic, dark azure and red, witchcore"' \ -i 'scheduler="K_EULER"' \ -i 'num_outputs=1' \ -i 'guidance_scale=7' \ -i 'apply_watermark=true' \ -i 'negative_prompt="(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31"' \ -i 'num_inference_steps=24'
To learn more, take a look at the Cog documentation.
Run this to download the model and run it in your local environment:
docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "seed": 201961, "width": 1024, "height": 1024, "prompt": "illustration, anime, Monkey D. Luffy, One Piece, vintage, port city, poster, in the style of [james jean], deathcore, [john pitre], detailed atmospheric portraits, oil painting, anime, biopunk, horror art, dark and intricate, unique illustrations, precisionist art, airbrush, mystical landscape, intricate psychedelic, dark azure and red, witchcore", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2023-11-13T21:35:44.100641Z", "created_at": "2023-11-13T21:35:40.093228Z", "data_removed": false, "error": null, "id": "q77swwlbbeyhicvbp32cxf24dm", "input": { "seed": 201961, "width": 1024, "height": 1024, "prompt": "illustration, anime, Monkey D. Luffy, One Piece, vintage, port city, poster, in the style of [james jean], deathcore, [john pitre], detailed atmospheric portraits, oil painting, anime, biopunk, horror art, dark and intricate, unique illustrations, precisionist art, airbrush, mystical landscape, intricate psychedelic, dark azure and red, witchcore", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 }, "logs": "Using seed: 201961\nThe following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: [', witchcore']\nThe following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: [', witchcore']\n 0%| | 0/24 [00:00<?, ?it/s]\n 4%|▍ | 1/24 [00:00<00:02, 8.73it/s]\n 8%|▊ | 2/24 [00:00<00:02, 8.73it/s]\n 12%|█▎ | 3/24 [00:00<00:02, 8.73it/s]\n 17%|█▋ | 4/24 [00:00<00:02, 8.75it/s]\n 21%|██ | 5/24 [00:00<00:02, 8.76it/s]\n 25%|██▌ | 6/24 [00:00<00:02, 8.75it/s]\n 29%|██▉ | 7/24 [00:00<00:01, 8.73it/s]\n 33%|███▎ | 8/24 [00:00<00:01, 8.72it/s]\n 38%|███▊ | 9/24 [00:01<00:01, 8.72it/s]\n 42%|████▏ | 10/24 [00:01<00:01, 8.71it/s]\n 46%|████▌ | 11/24 [00:01<00:01, 8.70it/s]\n 50%|█████ | 12/24 [00:01<00:01, 8.70it/s]\n 54%|█████▍ | 13/24 [00:01<00:01, 8.70it/s]\n 58%|█████▊ | 14/24 [00:01<00:01, 8.71it/s]\n 62%|██████▎ | 15/24 [00:01<00:01, 8.71it/s]\n 67%|██████▋ | 16/24 [00:01<00:00, 8.70it/s]\n 71%|███████ | 17/24 [00:01<00:00, 8.71it/s]\n 75%|███████▌ | 18/24 [00:02<00:00, 8.71it/s]\n 79%|███████▉ | 19/24 [00:02<00:00, 8.70it/s]\n 83%|████████▎ | 20/24 [00:02<00:00, 8.68it/s]\n 88%|████████▊ | 21/24 [00:02<00:00, 8.68it/s]\n 92%|█████████▏| 22/24 [00:02<00:00, 8.67it/s]\n 96%|█████████▌| 23/24 [00:02<00:00, 8.69it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.69it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.71it/s]", "metrics": { "predict_time": 4.119362, "total_time": 4.007413 }, "output": [ "https://replicate.delivery/pbxt/QK0BJQ5EzJ4YFZDXk5nlGWGCNgMZngaeY6q8H2dAvf2viE4RA/out-0.png" ], "started_at": "2023-11-13T21:35:39.981279Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/q77swwlbbeyhicvbp32cxf24dm", "cancel": "https://api.replicate.com/v1/predictions/q77swwlbbeyhicvbp32cxf24dm/cancel" }, "version": "9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c" }
Generated inUsing seed: 201961 The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: [', witchcore'] The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: [', witchcore'] 0%| | 0/24 [00:00<?, ?it/s] 4%|▍ | 1/24 [00:00<00:02, 8.73it/s] 8%|▊ | 2/24 [00:00<00:02, 8.73it/s] 12%|█▎ | 3/24 [00:00<00:02, 8.73it/s] 17%|█▋ | 4/24 [00:00<00:02, 8.75it/s] 21%|██ | 5/24 [00:00<00:02, 8.76it/s] 25%|██▌ | 6/24 [00:00<00:02, 8.75it/s] 29%|██▉ | 7/24 [00:00<00:01, 8.73it/s] 33%|███▎ | 8/24 [00:00<00:01, 8.72it/s] 38%|███▊ | 9/24 [00:01<00:01, 8.72it/s] 42%|████▏ | 10/24 [00:01<00:01, 8.71it/s] 46%|████▌ | 11/24 [00:01<00:01, 8.70it/s] 50%|█████ | 12/24 [00:01<00:01, 8.70it/s] 54%|█████▍ | 13/24 [00:01<00:01, 8.70it/s] 58%|█████▊ | 14/24 [00:01<00:01, 8.71it/s] 62%|██████▎ | 15/24 [00:01<00:01, 8.71it/s] 67%|██████▋ | 16/24 [00:01<00:00, 8.70it/s] 71%|███████ | 17/24 [00:01<00:00, 8.71it/s] 75%|███████▌ | 18/24 [00:02<00:00, 8.71it/s] 79%|███████▉ | 19/24 [00:02<00:00, 8.70it/s] 83%|████████▎ | 20/24 [00:02<00:00, 8.68it/s] 88%|████████▊ | 21/24 [00:02<00:00, 8.68it/s] 92%|█████████▏| 22/24 [00:02<00:00, 8.67it/s] 96%|█████████▌| 23/24 [00:02<00:00, 8.69it/s] 100%|██████████| 24/24 [00:02<00:00, 8.69it/s] 100%|██████████| 24/24 [00:02<00:00, 8.71it/s]
Prediction
lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449cIDsj26psdbxvscolzg4unkems5bqStatusSucceededSourceWebHardwareA100 (40GB)Total durationCreatedInput
- seed
- 615798744
- width
- 1024
- height
- 1024
- prompt
- Fairy, HDR, ultra high-detail, centered, bokeh, lens flare, foggy dusk, bloom, light bloom, cinematic lighting, high dynamic range, depth of field, full pose
- scheduler
- K_EULER
- num_outputs
- 1
- guidance_scale
- 7
- apply_watermark
- negative_prompt
- (worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1)
- num_inference_steps
- 24
{ "seed": 615798744, "width": 1024, "height": 1024, "prompt": "Fairy, HDR, ultra high-detail, centered, bokeh, lens flare, foggy dusk, bloom, light bloom, cinematic lighting, high dynamic range, depth of field, full pose", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1)", "num_inference_steps": 24 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; import fs from "node:fs"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", { input: { seed: 615798744, width: 1024, height: 1024, prompt: "Fairy, HDR, ultra high-detail, centered, bokeh, lens flare, foggy dusk, bloom, light bloom, cinematic lighting, high dynamic range, depth of field, full pose", scheduler: "K_EULER", num_outputs: 1, guidance_scale: 7, apply_watermark: true, negative_prompt: "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1)", num_inference_steps: 24 } } ); // To access the file URL: console.log(output[0].url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output[0]);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", input={ "seed": 615798744, "width": 1024, "height": 1024, "prompt": "Fairy, HDR, ultra high-detail, centered, bokeh, lens flare, foggy dusk, bloom, light bloom, cinematic lighting, high dynamic range, depth of field, full pose", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": True, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1)", "num_inference_steps": 24 } ) # To access the file URL: print(output[0].url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output[0].read())
To learn more, take a look at the guide on getting started with Python.
Run lucataco/sdxl-niji-se 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": "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", "input": { "seed": 615798744, "width": 1024, "height": 1024, "prompt": "Fairy, HDR, ultra high-detail, centered, bokeh, lens flare, foggy dusk, bloom, light bloom, cinematic lighting, high dynamic range, depth of field, full pose", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1)", "num_inference_steps": 24 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you don’t have Homebrew, there are other installation options available.
Run this to download the model and run it in your local environment:
cog predict r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c \ -i 'seed=615798744' \ -i 'width=1024' \ -i 'height=1024' \ -i 'prompt="Fairy, HDR, ultra high-detail, centered, bokeh, lens flare, foggy dusk, bloom, light bloom, cinematic lighting, high dynamic range, depth of field, full pose"' \ -i 'scheduler="K_EULER"' \ -i 'num_outputs=1' \ -i 'guidance_scale=7' \ -i 'apply_watermark=true' \ -i 'negative_prompt="(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1)"' \ -i 'num_inference_steps=24'
To learn more, take a look at the Cog documentation.
Run this to download the model and run it in your local environment:
docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "seed": 615798744, "width": 1024, "height": 1024, "prompt": "Fairy, HDR, ultra high-detail, centered, bokeh, lens flare, foggy dusk, bloom, light bloom, cinematic lighting, high dynamic range, depth of field, full pose", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1)", "num_inference_steps": 24 } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2023-11-13T21:36:15.076405Z", "created_at": "2023-11-13T21:36:11.071480Z", "data_removed": false, "error": null, "id": "sj26psdbxvscolzg4unkems5bq", "input": { "seed": 615798744, "width": 1024, "height": 1024, "prompt": "Fairy, HDR, ultra high-detail, centered, bokeh, lens flare, foggy dusk, bloom, light bloom, cinematic lighting, high dynamic range, depth of field, full pose", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1)", "num_inference_steps": 24 }, "logs": "Using seed: 615798744\n 0%| | 0/24 [00:00<?, ?it/s]\n 4%|▍ | 1/24 [00:00<00:02, 8.75it/s]\n 8%|▊ | 2/24 [00:00<00:02, 8.71it/s]\n 12%|█▎ | 3/24 [00:00<00:02, 8.72it/s]\n 17%|█▋ | 4/24 [00:00<00:02, 8.74it/s]\n 21%|██ | 5/24 [00:00<00:02, 8.74it/s]\n 25%|██▌ | 6/24 [00:00<00:02, 8.75it/s]\n 29%|██▉ | 7/24 [00:00<00:01, 8.75it/s]\n 33%|███▎ | 8/24 [00:00<00:01, 8.76it/s]\n 38%|███▊ | 9/24 [00:01<00:01, 8.74it/s]\n 42%|████▏ | 10/24 [00:01<00:01, 8.73it/s]\n 46%|████▌ | 11/24 [00:01<00:01, 8.75it/s]\n 50%|█████ | 12/24 [00:01<00:01, 8.73it/s]\n 54%|█████▍ | 13/24 [00:01<00:01, 8.72it/s]\n 58%|█████▊ | 14/24 [00:01<00:01, 8.71it/s]\n 62%|██████▎ | 15/24 [00:01<00:01, 8.70it/s]\n 67%|██████▋ | 16/24 [00:01<00:00, 8.71it/s]\n 71%|███████ | 17/24 [00:01<00:00, 8.71it/s]\n 75%|███████▌ | 18/24 [00:02<00:00, 8.71it/s]\n 79%|███████▉ | 19/24 [00:02<00:00, 8.71it/s]\n 83%|████████▎ | 20/24 [00:02<00:00, 8.71it/s]\n 88%|████████▊ | 21/24 [00:02<00:00, 8.71it/s]\n 92%|█████████▏| 22/24 [00:02<00:00, 8.72it/s]\n 96%|█████████▌| 23/24 [00:02<00:00, 8.72it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.72it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.72it/s]", "metrics": { "predict_time": 4.096417, "total_time": 4.004925 }, "output": [ "https://replicate.delivery/pbxt/quN3pcigOpLzLxtfy1POHRF3kMYrfFO46h0NFlB41PfcGJwjA/out-0.png" ], "started_at": "2023-11-13T21:36:10.979988Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/sj26psdbxvscolzg4unkems5bq", "cancel": "https://api.replicate.com/v1/predictions/sj26psdbxvscolzg4unkems5bq/cancel" }, "version": "9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c" }
Generated inUsing seed: 615798744 0%| | 0/24 [00:00<?, ?it/s] 4%|▍ | 1/24 [00:00<00:02, 8.75it/s] 8%|▊ | 2/24 [00:00<00:02, 8.71it/s] 12%|█▎ | 3/24 [00:00<00:02, 8.72it/s] 17%|█▋ | 4/24 [00:00<00:02, 8.74it/s] 21%|██ | 5/24 [00:00<00:02, 8.74it/s] 25%|██▌ | 6/24 [00:00<00:02, 8.75it/s] 29%|██▉ | 7/24 [00:00<00:01, 8.75it/s] 33%|███▎ | 8/24 [00:00<00:01, 8.76it/s] 38%|███▊ | 9/24 [00:01<00:01, 8.74it/s] 42%|████▏ | 10/24 [00:01<00:01, 8.73it/s] 46%|████▌ | 11/24 [00:01<00:01, 8.75it/s] 50%|█████ | 12/24 [00:01<00:01, 8.73it/s] 54%|█████▍ | 13/24 [00:01<00:01, 8.72it/s] 58%|█████▊ | 14/24 [00:01<00:01, 8.71it/s] 62%|██████▎ | 15/24 [00:01<00:01, 8.70it/s] 67%|██████▋ | 16/24 [00:01<00:00, 8.71it/s] 71%|███████ | 17/24 [00:01<00:00, 8.71it/s] 75%|███████▌ | 18/24 [00:02<00:00, 8.71it/s] 79%|███████▉ | 19/24 [00:02<00:00, 8.71it/s] 83%|████████▎ | 20/24 [00:02<00:00, 8.71it/s] 88%|████████▊ | 21/24 [00:02<00:00, 8.71it/s] 92%|█████████▏| 22/24 [00:02<00:00, 8.72it/s] 96%|█████████▌| 23/24 [00:02<00:00, 8.72it/s] 100%|██████████| 24/24 [00:02<00:00, 8.72it/s] 100%|██████████| 24/24 [00:02<00:00, 8.72it/s]
Prediction
lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449cIDve65yslbozcl5ij6j23mbfy5iiStatusSucceededSourceWebHardwareA100 (40GB)Total durationCreatedInput
- seed
- 1621100366
- width
- 1024
- height
- 1024
- prompt
- 1girl, gorgeous, delicate, blue hair, wearing light blue dress, barefoot, looking up into sky, standing in sea, perfect face, beautiful eyes, sky is galaxy and stars, twilight, shooting meteorites, longing
- scheduler
- K_EULER
- num_outputs
- 1
- guidance_scale
- 7
- apply_watermark
- negative_prompt
- (worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), unaestheticXLv31
- num_inference_steps
- 24
{ "seed": 1621100366, "width": 1024, "height": 1024, "prompt": "1girl, gorgeous, delicate, blue hair, wearing light blue dress, barefoot, looking up into sky, standing in sea, perfect face, beautiful eyes, sky is galaxy and stars, twilight, shooting meteorites, longing", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), unaestheticXLv31", "num_inference_steps": 24 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; import fs from "node:fs"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", { input: { seed: 1621100366, width: 1024, height: 1024, prompt: "1girl, gorgeous, delicate, blue hair, wearing light blue dress, barefoot, looking up into sky, standing in sea, perfect face, beautiful eyes, sky is galaxy and stars, twilight, shooting meteorites, longing", scheduler: "K_EULER", num_outputs: 1, guidance_scale: 7, apply_watermark: true, negative_prompt: "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), unaestheticXLv31", num_inference_steps: 24 } } ); // To access the file URL: console.log(output[0].url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output[0]);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", input={ "seed": 1621100366, "width": 1024, "height": 1024, "prompt": "1girl, gorgeous, delicate, blue hair, wearing light blue dress, barefoot, looking up into sky, standing in sea, perfect face, beautiful eyes, sky is galaxy and stars, twilight, shooting meteorites, longing", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": True, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), unaestheticXLv31", "num_inference_steps": 24 } ) # To access the file URL: print(output[0].url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output[0].read())
To learn more, take a look at the guide on getting started with Python.
Run lucataco/sdxl-niji-se 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": "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", "input": { "seed": 1621100366, "width": 1024, "height": 1024, "prompt": "1girl, gorgeous, delicate, blue hair, wearing light blue dress, barefoot, looking up into sky, standing in sea, perfect face, beautiful eyes, sky is galaxy and stars, twilight, shooting meteorites, longing", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), unaestheticXLv31", "num_inference_steps": 24 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you don’t have Homebrew, there are other installation options available.
Run this to download the model and run it in your local environment:
cog predict r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c \ -i 'seed=1621100366' \ -i 'width=1024' \ -i 'height=1024' \ -i 'prompt="1girl, gorgeous, delicate, blue hair, wearing light blue dress, barefoot, looking up into sky, standing in sea, perfect face, beautiful eyes, sky is galaxy and stars, twilight, shooting meteorites, longing"' \ -i 'scheduler="K_EULER"' \ -i 'num_outputs=1' \ -i 'guidance_scale=7' \ -i 'apply_watermark=true' \ -i 'negative_prompt="(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), unaestheticXLv31"' \ -i 'num_inference_steps=24'
To learn more, take a look at the Cog documentation.
Run this to download the model and run it in your local environment:
docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "seed": 1621100366, "width": 1024, "height": 1024, "prompt": "1girl, gorgeous, delicate, blue hair, wearing light blue dress, barefoot, looking up into sky, standing in sea, perfect face, beautiful eyes, sky is galaxy and stars, twilight, shooting meteorites, longing", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), unaestheticXLv31", "num_inference_steps": 24 } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2023-11-13T21:36:45.164495Z", "created_at": "2023-11-13T21:36:41.069176Z", "data_removed": false, "error": null, "id": "ve65yslbozcl5ij6j23mbfy5ii", "input": { "seed": 1621100366, "width": 1024, "height": 1024, "prompt": "1girl, gorgeous, delicate, blue hair, wearing light blue dress, barefoot, looking up into sky, standing in sea, perfect face, beautiful eyes, sky is galaxy and stars, twilight, shooting meteorites, longing", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), unaestheticXLv31", "num_inference_steps": 24 }, "logs": "Using seed: 1621100366\n 0%| | 0/24 [00:00<?, ?it/s]\n 4%|▍ | 1/24 [00:00<00:02, 8.74it/s]\n 8%|▊ | 2/24 [00:00<00:02, 8.72it/s]\n 12%|█▎ | 3/24 [00:00<00:02, 8.72it/s]\n 17%|█▋ | 4/24 [00:00<00:02, 8.72it/s]\n 21%|██ | 5/24 [00:00<00:02, 8.72it/s]\n 25%|██▌ | 6/24 [00:00<00:02, 8.71it/s]\n 29%|██▉ | 7/24 [00:00<00:01, 8.73it/s]\n 33%|███▎ | 8/24 [00:00<00:01, 8.73it/s]\n 38%|███▊ | 9/24 [00:01<00:01, 8.72it/s]\n 42%|████▏ | 10/24 [00:01<00:01, 8.72it/s]\n 46%|████▌ | 11/24 [00:01<00:01, 8.72it/s]\n 50%|█████ | 12/24 [00:01<00:01, 8.74it/s]\n 54%|█████▍ | 13/24 [00:01<00:01, 8.75it/s]\n 58%|█████▊ | 14/24 [00:01<00:01, 8.76it/s]\n 62%|██████▎ | 15/24 [00:01<00:01, 8.75it/s]\n 67%|██████▋ | 16/24 [00:01<00:00, 8.75it/s]\n 71%|███████ | 17/24 [00:01<00:00, 8.74it/s]\n 75%|███████▌ | 18/24 [00:02<00:00, 8.73it/s]\n 79%|███████▉ | 19/24 [00:02<00:00, 8.72it/s]\n 83%|████████▎ | 20/24 [00:02<00:00, 8.72it/s]\n 88%|████████▊ | 21/24 [00:02<00:00, 8.72it/s]\n 92%|█████████▏| 22/24 [00:02<00:00, 8.72it/s]\n 96%|█████████▌| 23/24 [00:02<00:00, 8.71it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.72it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.73it/s]", "metrics": { "predict_time": 4.119736, "total_time": 4.095319 }, "output": [ "https://replicate.delivery/pbxt/1ImkE1sCnL4SPtu6jLjxeimJneuNfffYE7Ib6GhzwhOmdkAPC/out-0.png" ], "started_at": "2023-11-13T21:36:41.044759Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/ve65yslbozcl5ij6j23mbfy5ii", "cancel": "https://api.replicate.com/v1/predictions/ve65yslbozcl5ij6j23mbfy5ii/cancel" }, "version": "9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c" }
Generated inUsing seed: 1621100366 0%| | 0/24 [00:00<?, ?it/s] 4%|▍ | 1/24 [00:00<00:02, 8.74it/s] 8%|▊ | 2/24 [00:00<00:02, 8.72it/s] 12%|█▎ | 3/24 [00:00<00:02, 8.72it/s] 17%|█▋ | 4/24 [00:00<00:02, 8.72it/s] 21%|██ | 5/24 [00:00<00:02, 8.72it/s] 25%|██▌ | 6/24 [00:00<00:02, 8.71it/s] 29%|██▉ | 7/24 [00:00<00:01, 8.73it/s] 33%|███▎ | 8/24 [00:00<00:01, 8.73it/s] 38%|███▊ | 9/24 [00:01<00:01, 8.72it/s] 42%|████▏ | 10/24 [00:01<00:01, 8.72it/s] 46%|████▌ | 11/24 [00:01<00:01, 8.72it/s] 50%|█████ | 12/24 [00:01<00:01, 8.74it/s] 54%|█████▍ | 13/24 [00:01<00:01, 8.75it/s] 58%|█████▊ | 14/24 [00:01<00:01, 8.76it/s] 62%|██████▎ | 15/24 [00:01<00:01, 8.75it/s] 67%|██████▋ | 16/24 [00:01<00:00, 8.75it/s] 71%|███████ | 17/24 [00:01<00:00, 8.74it/s] 75%|███████▌ | 18/24 [00:02<00:00, 8.73it/s] 79%|███████▉ | 19/24 [00:02<00:00, 8.72it/s] 83%|████████▎ | 20/24 [00:02<00:00, 8.72it/s] 88%|████████▊ | 21/24 [00:02<00:00, 8.72it/s] 92%|█████████▏| 22/24 [00:02<00:00, 8.72it/s] 96%|█████████▌| 23/24 [00:02<00:00, 8.71it/s] 100%|██████████| 24/24 [00:02<00:00, 8.72it/s] 100%|██████████| 24/24 [00:02<00:00, 8.73it/s]
Prediction
lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449cIDi4qxjp3b6p6cd4pu7bd22j5dkqStatusSucceededSourceWebHardwareA100 (40GB)Total durationCreatedInput
- seed
- 1606459437
- width
- 1024
- height
- 1024
- prompt
- strybk, 3d, bright pink sun, coral cloud formed spiral. tunnel effect, sunrise, oil painting, heavy strokes, surreal, paint dripping, digital art, concept art, surrealistic anime style, kawaii, octane render, hyperdetailed, magic, fantasy, fairy. Genshin Impact, Neimy Kanani, raphael, makoto shinkai style, disney, Lois van Baarle (Loish), Ilya Kuvshinov, Ross Tran (aka Rossdraws), Studio Ghibli, James Gilleard, artstation, 4k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style
- scheduler
- K_EULER
- num_outputs
- 1
- guidance_scale
- 7
- apply_watermark
- negative_prompt
- (worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31
- num_inference_steps
- 24
{ "seed": 1606459437, "width": 1024, "height": 1024, "prompt": "strybk, 3d, bright pink sun, coral cloud formed spiral. tunnel effect, sunrise, oil painting, heavy strokes, surreal, paint dripping, digital art, concept art, surrealistic anime style, kawaii, octane render, hyperdetailed, magic, fantasy, fairy. Genshin Impact, Neimy Kanani, raphael, makoto shinkai style, disney, Lois van Baarle (Loish), Ilya Kuvshinov, Ross Tran (aka Rossdraws), Studio Ghibli, James Gilleard, artstation, 4k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; import fs from "node:fs"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", { input: { seed: 1606459437, width: 1024, height: 1024, prompt: "strybk, 3d, bright pink sun, coral cloud formed spiral. tunnel effect, sunrise, oil painting, heavy strokes, surreal, paint dripping, digital art, concept art, surrealistic anime style, kawaii, octane render, hyperdetailed, magic, fantasy, fairy. Genshin Impact, Neimy Kanani, raphael, makoto shinkai style, disney, Lois van Baarle (Loish), Ilya Kuvshinov, Ross Tran (aka Rossdraws), Studio Ghibli, James Gilleard, artstation, 4k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style", scheduler: "K_EULER", num_outputs: 1, guidance_scale: 7, apply_watermark: true, negative_prompt: "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", num_inference_steps: 24 } } ); // To access the file URL: console.log(output[0].url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output[0]);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", input={ "seed": 1606459437, "width": 1024, "height": 1024, "prompt": "strybk, 3d, bright pink sun, coral cloud formed spiral. tunnel effect, sunrise, oil painting, heavy strokes, surreal, paint dripping, digital art, concept art, surrealistic anime style, kawaii, octane render, hyperdetailed, magic, fantasy, fairy. Genshin Impact, Neimy Kanani, raphael, makoto shinkai style, disney, Lois van Baarle (Loish), Ilya Kuvshinov, Ross Tran (aka Rossdraws), Studio Ghibli, James Gilleard, artstation, 4k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": True, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 } ) # To access the file URL: print(output[0].url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output[0].read())
To learn more, take a look at the guide on getting started with Python.
Run lucataco/sdxl-niji-se 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": "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", "input": { "seed": 1606459437, "width": 1024, "height": 1024, "prompt": "strybk, 3d, bright pink sun, coral cloud formed spiral. tunnel effect, sunrise, oil painting, heavy strokes, surreal, paint dripping, digital art, concept art, surrealistic anime style, kawaii, octane render, hyperdetailed, magic, fantasy, fairy. Genshin Impact, Neimy Kanani, raphael, makoto shinkai style, disney, Lois van Baarle (Loish), Ilya Kuvshinov, Ross Tran (aka Rossdraws), Studio Ghibli, James Gilleard, artstation, 4k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you don’t have Homebrew, there are other installation options available.
Run this to download the model and run it in your local environment:
cog predict r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c \ -i 'seed=1606459437' \ -i 'width=1024' \ -i 'height=1024' \ -i 'prompt="strybk, 3d, bright pink sun, coral cloud formed spiral. tunnel effect, sunrise, oil painting, heavy strokes, surreal, paint dripping, digital art, concept art, surrealistic anime style, kawaii, octane render, hyperdetailed, magic, fantasy, fairy. Genshin Impact, Neimy Kanani, raphael, makoto shinkai style, disney, Lois van Baarle (Loish), Ilya Kuvshinov, Ross Tran (aka Rossdraws), Studio Ghibli, James Gilleard, artstation, 4k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style"' \ -i 'scheduler="K_EULER"' \ -i 'num_outputs=1' \ -i 'guidance_scale=7' \ -i 'apply_watermark=true' \ -i 'negative_prompt="(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31"' \ -i 'num_inference_steps=24'
To learn more, take a look at the Cog documentation.
Run this to download the model and run it in your local environment:
docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "seed": 1606459437, "width": 1024, "height": 1024, "prompt": "strybk, 3d, bright pink sun, coral cloud formed spiral. tunnel effect, sunrise, oil painting, heavy strokes, surreal, paint dripping, digital art, concept art, surrealistic anime style, kawaii, octane render, hyperdetailed, magic, fantasy, fairy. Genshin Impact, Neimy Kanani, raphael, makoto shinkai style, disney, Lois van Baarle (Loish), Ilya Kuvshinov, Ross Tran (aka Rossdraws), Studio Ghibli, James Gilleard, artstation, 4k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2023-11-13T21:37:22.740576Z", "created_at": "2023-11-13T21:37:18.697983Z", "data_removed": false, "error": null, "id": "i4qxjp3b6p6cd4pu7bd22j5dkq", "input": { "seed": 1606459437, "width": 1024, "height": 1024, "prompt": "strybk, 3d, bright pink sun, coral cloud formed spiral. tunnel effect, sunrise, oil painting, heavy strokes, surreal, paint dripping, digital art, concept art, surrealistic anime style, kawaii, octane render, hyperdetailed, magic, fantasy, fairy. Genshin Impact, Neimy Kanani, raphael, makoto shinkai style, disney, Lois van Baarle (Loish), Ilya Kuvshinov, Ross Tran (aka Rossdraws), Studio Ghibli, James Gilleard, artstation, 4k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 }, "logs": "Using seed: 1606459437\nThe following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['lois van baarle ( loish ), ilya kuvshinov, ross tran ( aka rossdraws ), studio ghibli, james gilleard, artstation, 4 k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style']\nThe following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['lois van baarle ( loish ), ilya kuvshinov, ross tran ( aka rossdraws ), studio ghibli, james gilleard, artstation, 4 k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style']\n 0%| | 0/24 [00:00<?, ?it/s]\n 4%|▍ | 1/24 [00:00<00:02, 8.76it/s]\n 8%|▊ | 2/24 [00:00<00:02, 8.72it/s]\n 12%|█▎ | 3/24 [00:00<00:02, 8.75it/s]\n 17%|█▋ | 4/24 [00:00<00:02, 8.76it/s]\n 21%|██ | 5/24 [00:00<00:02, 8.77it/s]\n 25%|██▌ | 6/24 [00:00<00:02, 8.75it/s]\n 29%|██▉ | 7/24 [00:00<00:01, 8.77it/s]\n 33%|███▎ | 8/24 [00:00<00:01, 8.77it/s]\n 38%|███▊ | 9/24 [00:01<00:01, 8.77it/s]\n 42%|████▏ | 10/24 [00:01<00:01, 8.77it/s]\n 46%|████▌ | 11/24 [00:01<00:01, 8.76it/s]\n 50%|█████ | 12/24 [00:01<00:01, 8.76it/s]\n 54%|█████▍ | 13/24 [00:01<00:01, 8.75it/s]\n 58%|█████▊ | 14/24 [00:01<00:01, 8.74it/s]\n 62%|██████▎ | 15/24 [00:01<00:01, 8.74it/s]\n 67%|██████▋ | 16/24 [00:01<00:00, 8.74it/s]\n 71%|███████ | 17/24 [00:01<00:00, 8.75it/s]\n 75%|███████▌ | 18/24 [00:02<00:00, 8.75it/s]\n 79%|███████▉ | 19/24 [00:02<00:00, 8.74it/s]\n 83%|████████▎ | 20/24 [00:02<00:00, 8.73it/s]\n 88%|████████▊ | 21/24 [00:02<00:00, 8.73it/s]\n 92%|█████████▏| 22/24 [00:02<00:00, 8.72it/s]\n 96%|█████████▌| 23/24 [00:02<00:00, 8.73it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.72it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.74it/s]", "metrics": { "predict_time": 4.119513, "total_time": 4.042593 }, "output": [ "https://replicate.delivery/pbxt/2vOXsaLqmQ6qP15rGdPeVveHfpjQg2sW9njNJY4SDajlIJwjA/out-0.png" ], "started_at": "2023-11-13T21:37:18.621063Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/i4qxjp3b6p6cd4pu7bd22j5dkq", "cancel": "https://api.replicate.com/v1/predictions/i4qxjp3b6p6cd4pu7bd22j5dkq/cancel" }, "version": "9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c" }
Generated inUsing seed: 1606459437 The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['lois van baarle ( loish ), ilya kuvshinov, ross tran ( aka rossdraws ), studio ghibli, james gilleard, artstation, 4 k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style'] The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['lois van baarle ( loish ), ilya kuvshinov, ross tran ( aka rossdraws ), studio ghibli, james gilleard, artstation, 4 k, symmetrical, soft lighting, trending on artstation, intricate details, highly detailed, pencil drawing, sketching, unreal engine, kids story book style, muted colors, watercolor style'] 0%| | 0/24 [00:00<?, ?it/s] 4%|▍ | 1/24 [00:00<00:02, 8.76it/s] 8%|▊ | 2/24 [00:00<00:02, 8.72it/s] 12%|█▎ | 3/24 [00:00<00:02, 8.75it/s] 17%|█▋ | 4/24 [00:00<00:02, 8.76it/s] 21%|██ | 5/24 [00:00<00:02, 8.77it/s] 25%|██▌ | 6/24 [00:00<00:02, 8.75it/s] 29%|██▉ | 7/24 [00:00<00:01, 8.77it/s] 33%|███▎ | 8/24 [00:00<00:01, 8.77it/s] 38%|███▊ | 9/24 [00:01<00:01, 8.77it/s] 42%|████▏ | 10/24 [00:01<00:01, 8.77it/s] 46%|████▌ | 11/24 [00:01<00:01, 8.76it/s] 50%|█████ | 12/24 [00:01<00:01, 8.76it/s] 54%|█████▍ | 13/24 [00:01<00:01, 8.75it/s] 58%|█████▊ | 14/24 [00:01<00:01, 8.74it/s] 62%|██████▎ | 15/24 [00:01<00:01, 8.74it/s] 67%|██████▋ | 16/24 [00:01<00:00, 8.74it/s] 71%|███████ | 17/24 [00:01<00:00, 8.75it/s] 75%|███████▌ | 18/24 [00:02<00:00, 8.75it/s] 79%|███████▉ | 19/24 [00:02<00:00, 8.74it/s] 83%|████████▎ | 20/24 [00:02<00:00, 8.73it/s] 88%|████████▊ | 21/24 [00:02<00:00, 8.73it/s] 92%|█████████▏| 22/24 [00:02<00:00, 8.72it/s] 96%|█████████▌| 23/24 [00:02<00:00, 8.73it/s] 100%|██████████| 24/24 [00:02<00:00, 8.72it/s] 100%|██████████| 24/24 [00:02<00:00, 8.74it/s]
Prediction
lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449cIDncybeslbxnlmaxgdmjwjitqpsqStatusSucceededSourceWebHardwareA100 (40GB)Total durationCreatedInput
- seed
- 377612558
- width
- 1024
- height
- 1024
- prompt
- 1980s anime, adult male, red haired bounty hunter in gothic city, distinct characterization, 2D animation aesthetic, muted pastel colors, in style of Yoshihiro Togashi
- scheduler
- K_EULER
- num_outputs
- 1
- guidance_scale
- 7
- apply_watermark
- negative_prompt
- (worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31
- num_inference_steps
- 24
{ "seed": 377612558, "width": 1024, "height": 1024, "prompt": "1980s anime, adult male, red haired bounty hunter in gothic city, distinct characterization, 2D animation aesthetic, muted pastel colors, in style of Yoshihiro Togashi", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; import fs from "node:fs"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", { input: { seed: 377612558, width: 1024, height: 1024, prompt: "1980s anime, adult male, red haired bounty hunter in gothic city, distinct characterization, 2D animation aesthetic, muted pastel colors, in style of Yoshihiro Togashi", scheduler: "K_EULER", num_outputs: 1, guidance_scale: 7, apply_watermark: true, negative_prompt: "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", num_inference_steps: 24 } } ); // To access the file URL: console.log(output[0].url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output[0]);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", input={ "seed": 377612558, "width": 1024, "height": 1024, "prompt": "1980s anime, adult male, red haired bounty hunter in gothic city, distinct characterization, 2D animation aesthetic, muted pastel colors, in style of Yoshihiro Togashi", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": True, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 } ) # To access the file URL: print(output[0].url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output[0].read())
To learn more, take a look at the guide on getting started with Python.
Run lucataco/sdxl-niji-se 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": "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", "input": { "seed": 377612558, "width": 1024, "height": 1024, "prompt": "1980s anime, adult male, red haired bounty hunter in gothic city, distinct characterization, 2D animation aesthetic, muted pastel colors, in style of Yoshihiro Togashi", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you don’t have Homebrew, there are other installation options available.
Run this to download the model and run it in your local environment:
cog predict r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c \ -i 'seed=377612558' \ -i 'width=1024' \ -i 'height=1024' \ -i 'prompt="1980s anime, adult male, red haired bounty hunter in gothic city, distinct characterization, 2D animation aesthetic, muted pastel colors, in style of Yoshihiro Togashi"' \ -i 'scheduler="K_EULER"' \ -i 'num_outputs=1' \ -i 'guidance_scale=7' \ -i 'apply_watermark=true' \ -i 'negative_prompt="(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31"' \ -i 'num_inference_steps=24'
To learn more, take a look at the Cog documentation.
Run this to download the model and run it in your local environment:
docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "seed": 377612558, "width": 1024, "height": 1024, "prompt": "1980s anime, adult male, red haired bounty hunter in gothic city, distinct characterization, 2D animation aesthetic, muted pastel colors, in style of Yoshihiro Togashi", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2023-11-13T22:20:47.647320Z", "created_at": "2023-11-13T22:17:45.479572Z", "data_removed": false, "error": null, "id": "ncybeslbxnlmaxgdmjwjitqpsq", "input": { "seed": 377612558, "width": 1024, "height": 1024, "prompt": "1980s anime, adult male, red haired bounty hunter in gothic city, distinct characterization, 2D animation aesthetic, muted pastel colors, in style of Yoshihiro Togashi", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "(worst quality, low quality, normal quality, bad quality:1.4), (lowres:1.3), (extra fingers, missing fingers:1.2), (poorly rendered hands:1.2), (mutation:1.1), (mutated hands:1.3), (deformed iris, deformed pupils:1.1), watermark, missing arms, missing hands, (long neck:1.4), (poorly drawn:1.3), (poorly drawn eyes:1.2), (poorly drawn hands:1.3), (poorly drawn iris:1.2), (poorly drawn pupils:1.2), (realistic:1.5), (blurry:1.1), unaestheticXLv31", "num_inference_steps": 24 }, "logs": "Using seed: 377612558\n 0%| | 0/24 [00:00<?, ?it/s]\n 4%|▍ | 1/24 [00:00<00:09, 2.34it/s]\n 8%|▊ | 2/24 [00:00<00:05, 4.12it/s]\n 12%|█▎ | 3/24 [00:00<00:03, 5.42it/s]\n 17%|█▋ | 4/24 [00:00<00:03, 6.36it/s]\n 21%|██ | 5/24 [00:00<00:02, 7.05it/s]\n 25%|██▌ | 6/24 [00:01<00:02, 7.54it/s]\n 29%|██▉ | 7/24 [00:01<00:02, 7.90it/s]\n 33%|███▎ | 8/24 [00:01<00:01, 8.14it/s]\n 38%|███▊ | 9/24 [00:01<00:01, 8.32it/s]\n 42%|████▏ | 10/24 [00:01<00:01, 8.44it/s]\n 46%|████▌ | 11/24 [00:01<00:01, 8.53it/s]\n 50%|█████ | 12/24 [00:01<00:01, 8.57it/s]\n 54%|█████▍ | 13/24 [00:01<00:01, 8.62it/s]\n 58%|█████▊ | 14/24 [00:01<00:01, 8.66it/s]\n 62%|██████▎ | 15/24 [00:02<00:01, 8.68it/s]\n 67%|██████▋ | 16/24 [00:02<00:00, 8.68it/s]\n 71%|███████ | 17/24 [00:02<00:00, 8.71it/s]\n 75%|███████▌ | 18/24 [00:02<00:00, 8.72it/s]\n 79%|███████▉ | 19/24 [00:02<00:00, 8.71it/s]\n 83%|████████▎ | 20/24 [00:02<00:00, 8.71it/s]\n 88%|████████▊ | 21/24 [00:02<00:00, 8.71it/s]\n 92%|█████████▏| 22/24 [00:02<00:00, 8.72it/s]\n 96%|█████████▌| 23/24 [00:02<00:00, 8.72it/s]\n100%|██████████| 24/24 [00:03<00:00, 8.71it/s]\n100%|██████████| 24/24 [00:03<00:00, 7.83it/s]", "metrics": { "predict_time": 4.993873, "total_time": 182.167748 }, "output": [ "https://replicate.delivery/pbxt/HR6WtFZjvL5FNNpbouXsDPbdpAwhQoY3WJoaQOjrwp0PTBeIA/out-0.png" ], "started_at": "2023-11-13T22:20:42.653447Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/ncybeslbxnlmaxgdmjwjitqpsq", "cancel": "https://api.replicate.com/v1/predictions/ncybeslbxnlmaxgdmjwjitqpsq/cancel" }, "version": "9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c" }
Generated inUsing seed: 377612558 0%| | 0/24 [00:00<?, ?it/s] 4%|▍ | 1/24 [00:00<00:09, 2.34it/s] 8%|▊ | 2/24 [00:00<00:05, 4.12it/s] 12%|█▎ | 3/24 [00:00<00:03, 5.42it/s] 17%|█▋ | 4/24 [00:00<00:03, 6.36it/s] 21%|██ | 5/24 [00:00<00:02, 7.05it/s] 25%|██▌ | 6/24 [00:01<00:02, 7.54it/s] 29%|██▉ | 7/24 [00:01<00:02, 7.90it/s] 33%|███▎ | 8/24 [00:01<00:01, 8.14it/s] 38%|███▊ | 9/24 [00:01<00:01, 8.32it/s] 42%|████▏ | 10/24 [00:01<00:01, 8.44it/s] 46%|████▌ | 11/24 [00:01<00:01, 8.53it/s] 50%|█████ | 12/24 [00:01<00:01, 8.57it/s] 54%|█████▍ | 13/24 [00:01<00:01, 8.62it/s] 58%|█████▊ | 14/24 [00:01<00:01, 8.66it/s] 62%|██████▎ | 15/24 [00:02<00:01, 8.68it/s] 67%|██████▋ | 16/24 [00:02<00:00, 8.68it/s] 71%|███████ | 17/24 [00:02<00:00, 8.71it/s] 75%|███████▌ | 18/24 [00:02<00:00, 8.72it/s] 79%|███████▉ | 19/24 [00:02<00:00, 8.71it/s] 83%|████████▎ | 20/24 [00:02<00:00, 8.71it/s] 88%|████████▊ | 21/24 [00:02<00:00, 8.71it/s] 92%|█████████▏| 22/24 [00:02<00:00, 8.72it/s] 96%|█████████▌| 23/24 [00:02<00:00, 8.72it/s] 100%|██████████| 24/24 [00:03<00:00, 8.71it/s] 100%|██████████| 24/24 [00:03<00:00, 7.83it/s]
Prediction
lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449cIDxv6za4dbuh27tcuhlzcikkuq6qStatusSucceededSourceWebHardwareA100 (40GB)Total durationCreatedInput
- seed
- 1673572517
- width
- 1024
- height
- 1024
- prompt
- anime, rpg, a chinese dragon spirit emerges from a waterfall opening a portal
- scheduler
- K_EULER
- num_outputs
- 1
- guidance_scale
- 7
- apply_watermark
- negative_prompt
- bad hands, bad anatomy
- num_inference_steps
- 24
{ "seed": 1673572517, "width": 1024, "height": 1024, "prompt": "anime, rpg, a chinese dragon spirit emerges from a waterfall opening a portal", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "bad hands, bad anatomy", "num_inference_steps": 24 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; import fs from "node:fs"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", { input: { seed: 1673572517, width: 1024, height: 1024, prompt: "anime, rpg, a chinese dragon spirit emerges from a waterfall opening a portal", scheduler: "K_EULER", num_outputs: 1, guidance_scale: 7, apply_watermark: true, negative_prompt: "bad hands, bad anatomy", num_inference_steps: 24 } } ); // To access the file URL: console.log(output[0].url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output[0]);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", input={ "seed": 1673572517, "width": 1024, "height": 1024, "prompt": "anime, rpg, a chinese dragon spirit emerges from a waterfall opening a portal", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": True, "negative_prompt": "bad hands, bad anatomy", "num_inference_steps": 24 } ) # To access the file URL: print(output[0].url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output[0].read())
To learn more, take a look at the guide on getting started with Python.
Run lucataco/sdxl-niji-se 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": "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", "input": { "seed": 1673572517, "width": 1024, "height": 1024, "prompt": "anime, rpg, a chinese dragon spirit emerges from a waterfall opening a portal", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "bad hands, bad anatomy", "num_inference_steps": 24 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you don’t have Homebrew, there are other installation options available.
Run this to download the model and run it in your local environment:
cog predict r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c \ -i 'seed=1673572517' \ -i 'width=1024' \ -i 'height=1024' \ -i 'prompt="anime, rpg, a chinese dragon spirit emerges from a waterfall opening a portal"' \ -i 'scheduler="K_EULER"' \ -i 'num_outputs=1' \ -i 'guidance_scale=7' \ -i 'apply_watermark=true' \ -i 'negative_prompt="bad hands, bad anatomy"' \ -i 'num_inference_steps=24'
To learn more, take a look at the Cog documentation.
Run this to download the model and run it in your local environment:
docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "seed": 1673572517, "width": 1024, "height": 1024, "prompt": "anime, rpg, a chinese dragon spirit emerges from a waterfall opening a portal", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "bad hands, bad anatomy", "num_inference_steps": 24 } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2023-11-13T22:21:47.057118Z", "created_at": "2023-11-13T22:21:42.948040Z", "data_removed": false, "error": null, "id": "xv6za4dbuh27tcuhlzcikkuq6q", "input": { "seed": 1673572517, "width": 1024, "height": 1024, "prompt": "anime, rpg, a chinese dragon spirit emerges from a waterfall opening a portal", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "bad hands, bad anatomy", "num_inference_steps": 24 }, "logs": "Using seed: 1673572517\n 0%| | 0/24 [00:00<?, ?it/s]\n 4%|▍ | 1/24 [00:00<00:02, 8.89it/s]\n 8%|▊ | 2/24 [00:00<00:02, 8.83it/s]\n 12%|█▎ | 3/24 [00:00<00:02, 8.78it/s]\n 17%|█▋ | 4/24 [00:00<00:02, 8.77it/s]\n 21%|██ | 5/24 [00:00<00:02, 8.77it/s]\n 25%|██▌ | 6/24 [00:00<00:02, 8.73it/s]\n 29%|██▉ | 7/24 [00:00<00:01, 8.73it/s]\n 33%|███▎ | 8/24 [00:00<00:01, 8.73it/s]\n 38%|███▊ | 9/24 [00:01<00:01, 8.72it/s]\n 42%|████▏ | 10/24 [00:01<00:01, 8.73it/s]\n 46%|████▌ | 11/24 [00:01<00:01, 8.74it/s]\n 50%|█████ | 12/24 [00:01<00:01, 8.73it/s]\n 54%|█████▍ | 13/24 [00:01<00:01, 8.72it/s]\n 58%|█████▊ | 14/24 [00:01<00:01, 8.72it/s]\n 62%|██████▎ | 15/24 [00:01<00:01, 8.72it/s]\n 67%|██████▋ | 16/24 [00:01<00:00, 8.71it/s]\n 71%|███████ | 17/24 [00:01<00:00, 8.71it/s]\n 75%|███████▌ | 18/24 [00:02<00:00, 8.71it/s]\n 79%|███████▉ | 19/24 [00:02<00:00, 8.70it/s]\n 83%|████████▎ | 20/24 [00:02<00:00, 8.70it/s]\n 88%|████████▊ | 21/24 [00:02<00:00, 8.70it/s]\n 92%|█████████▏| 22/24 [00:02<00:00, 8.71it/s]\n 96%|█████████▌| 23/24 [00:02<00:00, 8.73it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.74it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.73it/s]", "metrics": { "predict_time": 4.152654, "total_time": 4.109078 }, "output": [ "https://replicate.delivery/pbxt/l5rhAkJebqwefJ2FwxkkV6fg8VxePbwjw2rmFiYwCTMSvpAPC/out-0.png" ], "started_at": "2023-11-13T22:21:42.904464Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/xv6za4dbuh27tcuhlzcikkuq6q", "cancel": "https://api.replicate.com/v1/predictions/xv6za4dbuh27tcuhlzcikkuq6q/cancel" }, "version": "9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c" }
Generated inUsing seed: 1673572517 0%| | 0/24 [00:00<?, ?it/s] 4%|▍ | 1/24 [00:00<00:02, 8.89it/s] 8%|▊ | 2/24 [00:00<00:02, 8.83it/s] 12%|█▎ | 3/24 [00:00<00:02, 8.78it/s] 17%|█▋ | 4/24 [00:00<00:02, 8.77it/s] 21%|██ | 5/24 [00:00<00:02, 8.77it/s] 25%|██▌ | 6/24 [00:00<00:02, 8.73it/s] 29%|██▉ | 7/24 [00:00<00:01, 8.73it/s] 33%|███▎ | 8/24 [00:00<00:01, 8.73it/s] 38%|███▊ | 9/24 [00:01<00:01, 8.72it/s] 42%|████▏ | 10/24 [00:01<00:01, 8.73it/s] 46%|████▌ | 11/24 [00:01<00:01, 8.74it/s] 50%|█████ | 12/24 [00:01<00:01, 8.73it/s] 54%|█████▍ | 13/24 [00:01<00:01, 8.72it/s] 58%|█████▊ | 14/24 [00:01<00:01, 8.72it/s] 62%|██████▎ | 15/24 [00:01<00:01, 8.72it/s] 67%|██████▋ | 16/24 [00:01<00:00, 8.71it/s] 71%|███████ | 17/24 [00:01<00:00, 8.71it/s] 75%|███████▌ | 18/24 [00:02<00:00, 8.71it/s] 79%|███████▉ | 19/24 [00:02<00:00, 8.70it/s] 83%|████████▎ | 20/24 [00:02<00:00, 8.70it/s] 88%|████████▊ | 21/24 [00:02<00:00, 8.70it/s] 92%|█████████▏| 22/24 [00:02<00:00, 8.71it/s] 96%|█████████▌| 23/24 [00:02<00:00, 8.73it/s] 100%|██████████| 24/24 [00:02<00:00, 8.74it/s] 100%|██████████| 24/24 [00:02<00:00, 8.73it/s]
Prediction
lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449cIDvcumxjtbn76cgms7kudj6hkppuStatusSucceededSourceWebHardwareA100 (40GB)Total durationCreatedInput
- seed
- 298598707
- width
- 1024
- height
- 1024
- prompt
- Disney Alice in Wonderland, hacking on a computer. Large window, cyberpunk cityscape.
- scheduler
- K_EULER
- num_outputs
- 1
- guidance_scale
- 7
- apply_watermark
- negative_prompt
- Photo, deformed, black and white, realism, disfigured, low contrast
- num_inference_steps
- 24
{ "seed": 298598707, "width": 1024, "height": 1024, "prompt": "Disney Alice in Wonderland, hacking on a computer. Large window, cyberpunk cityscape.", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "Photo, deformed, black and white, realism, disfigured, low contrast", "num_inference_steps": 24 }
Install Replicate’s Node.js client library:npm install replicate
Import and set up the client:import Replicate from "replicate"; import fs from "node:fs"; const replicate = new Replicate({ auth: process.env.REPLICATE_API_TOKEN, });
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", { input: { seed: 298598707, width: 1024, height: 1024, prompt: "Disney Alice in Wonderland, hacking on a computer. Large window, cyberpunk cityscape.", scheduler: "K_EULER", num_outputs: 1, guidance_scale: 7, apply_watermark: true, negative_prompt: "Photo, deformed, black and white, realism, disfigured, low contrast", num_inference_steps: 24 } } ); // To access the file URL: console.log(output[0].url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output[0]);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run lucataco/sdxl-niji-se using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", input={ "seed": 298598707, "width": 1024, "height": 1024, "prompt": "Disney Alice in Wonderland, hacking on a computer. Large window, cyberpunk cityscape.", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": True, "negative_prompt": "Photo, deformed, black and white, realism, disfigured, low contrast", "num_inference_steps": 24 } ) # To access the file URL: print(output[0].url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output[0].read())
To learn more, take a look at the guide on getting started with Python.
Run lucataco/sdxl-niji-se 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": "lucataco/sdxl-niji-se:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c", "input": { "seed": 298598707, "width": 1024, "height": 1024, "prompt": "Disney Alice in Wonderland, hacking on a computer. Large window, cyberpunk cityscape.", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "Photo, deformed, black and white, realism, disfigured, low contrast", "num_inference_steps": 24 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you don’t have Homebrew, there are other installation options available.
Run this to download the model and run it in your local environment:
cog predict r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c \ -i 'seed=298598707' \ -i 'width=1024' \ -i 'height=1024' \ -i 'prompt="Disney Alice in Wonderland, hacking on a computer. Large window, cyberpunk cityscape."' \ -i 'scheduler="K_EULER"' \ -i 'num_outputs=1' \ -i 'guidance_scale=7' \ -i 'apply_watermark=true' \ -i 'negative_prompt="Photo, deformed, black and white, realism, disfigured, low contrast"' \ -i 'num_inference_steps=24'
To learn more, take a look at the Cog documentation.
Run this to download the model and run it in your local environment:
docker run -d -p 5000:5000 --gpus=all r8.im/lucataco/sdxl-niji-se@sha256:9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "seed": 298598707, "width": 1024, "height": 1024, "prompt": "Disney Alice in Wonderland, hacking on a computer. Large window, cyberpunk cityscape.", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "Photo, deformed, black and white, realism, disfigured, low contrast", "num_inference_steps": 24 } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2023-11-13T22:23:11.123504Z", "created_at": "2023-11-13T22:23:07.129872Z", "data_removed": false, "error": null, "id": "vcumxjtbn76cgms7kudj6hkppu", "input": { "seed": 298598707, "width": 1024, "height": 1024, "prompt": "Disney Alice in Wonderland, hacking on a computer. Large window, cyberpunk cityscape.", "scheduler": "K_EULER", "num_outputs": 1, "guidance_scale": 7, "apply_watermark": true, "negative_prompt": "Photo, deformed, black and white, realism, disfigured, low contrast", "num_inference_steps": 24 }, "logs": "Using seed: 298598707\n 0%| | 0/24 [00:00<?, ?it/s]\n 4%|▍ | 1/24 [00:00<00:02, 8.85it/s]\n 8%|▊ | 2/24 [00:00<00:02, 8.78it/s]\n 12%|█▎ | 3/24 [00:00<00:02, 8.79it/s]\n 17%|█▋ | 4/24 [00:00<00:02, 8.77it/s]\n 21%|██ | 5/24 [00:00<00:02, 8.77it/s]\n 25%|██▌ | 6/24 [00:00<00:02, 8.76it/s]\n 29%|██▉ | 7/24 [00:00<00:01, 8.75it/s]\n 33%|███▎ | 8/24 [00:00<00:01, 8.74it/s]\n 38%|███▊ | 9/24 [00:01<00:01, 8.73it/s]\n 42%|████▏ | 10/24 [00:01<00:01, 8.73it/s]\n 46%|████▌ | 11/24 [00:01<00:01, 8.73it/s]\n 50%|█████ | 12/24 [00:01<00:01, 8.73it/s]\n 54%|█████▍ | 13/24 [00:01<00:01, 8.73it/s]\n 58%|█████▊ | 14/24 [00:01<00:01, 8.72it/s]\n 62%|██████▎ | 15/24 [00:01<00:01, 8.71it/s]\n 67%|██████▋ | 16/24 [00:01<00:00, 8.72it/s]\n 71%|███████ | 17/24 [00:01<00:00, 8.72it/s]\n 75%|███████▌ | 18/24 [00:02<00:00, 8.72it/s]\n 79%|███████▉ | 19/24 [00:02<00:00, 8.71it/s]\n 83%|████████▎ | 20/24 [00:02<00:00, 8.70it/s]\n 88%|████████▊ | 21/24 [00:02<00:00, 8.69it/s]\n 92%|█████████▏| 22/24 [00:02<00:00, 8.70it/s]\n 96%|█████████▌| 23/24 [00:02<00:00, 8.71it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.70it/s]\n100%|██████████| 24/24 [00:02<00:00, 8.73it/s]", "metrics": { "predict_time": 4.036467, "total_time": 3.993632 }, "output": [ "https://replicate.delivery/pbxt/UJnXBfAh7d1tMiTqWWASV7dLTq5ZeFhAvvWDL96vA96OPF4RA/out-0.png" ], "started_at": "2023-11-13T22:23:07.087037Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/vcumxjtbn76cgms7kudj6hkppu", "cancel": "https://api.replicate.com/v1/predictions/vcumxjtbn76cgms7kudj6hkppu/cancel" }, "version": "9e0521b45e1774fc3288ec9ffa964d064c9d905e1ca50f37549084b021d7449c" }
Generated inUsing seed: 298598707 0%| | 0/24 [00:00<?, ?it/s] 4%|▍ | 1/24 [00:00<00:02, 8.85it/s] 8%|▊ | 2/24 [00:00<00:02, 8.78it/s] 12%|█▎ | 3/24 [00:00<00:02, 8.79it/s] 17%|█▋ | 4/24 [00:00<00:02, 8.77it/s] 21%|██ | 5/24 [00:00<00:02, 8.77it/s] 25%|██▌ | 6/24 [00:00<00:02, 8.76it/s] 29%|██▉ | 7/24 [00:00<00:01, 8.75it/s] 33%|███▎ | 8/24 [00:00<00:01, 8.74it/s] 38%|███▊ | 9/24 [00:01<00:01, 8.73it/s] 42%|████▏ | 10/24 [00:01<00:01, 8.73it/s] 46%|████▌ | 11/24 [00:01<00:01, 8.73it/s] 50%|█████ | 12/24 [00:01<00:01, 8.73it/s] 54%|█████▍ | 13/24 [00:01<00:01, 8.73it/s] 58%|█████▊ | 14/24 [00:01<00:01, 8.72it/s] 62%|██████▎ | 15/24 [00:01<00:01, 8.71it/s] 67%|██████▋ | 16/24 [00:01<00:00, 8.72it/s] 71%|███████ | 17/24 [00:01<00:00, 8.72it/s] 75%|███████▌ | 18/24 [00:02<00:00, 8.72it/s] 79%|███████▉ | 19/24 [00:02<00:00, 8.71it/s] 83%|████████▎ | 20/24 [00:02<00:00, 8.70it/s] 88%|████████▊ | 21/24 [00:02<00:00, 8.69it/s] 92%|█████████▏| 22/24 [00:02<00:00, 8.70it/s] 96%|█████████▌| 23/24 [00:02<00:00, 8.71it/s] 100%|██████████| 24/24 [00:02<00:00, 8.70it/s] 100%|██████████| 24/24 [00:02<00:00, 8.73it/s]
Want to make some of these yourself?
Run this model