zsxkib / sd3-controlnet
β¨Stable Diffusion 3 w/ β‘InstantX's Canny, Pose, and Tile ControlNetsπΌοΈ
Prediction
zsxkib/sd3-controlnet:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897IDr62c05e23xrgp0cg7b8ssxesmrStatusSucceededSourceWebHardwareA40 (Large)Total durationCreatedInput
- prompt
- vivid artistic style, women wearing a pinstripe suit
- structure
- tile
- output_format
- webp
- control_weight
- 0.3
- guidance_scale
- 7
- output_quality
- 80
- inference_steps
- 25
- negative_prompt
- NSFW, nude, naked, porn, ugly
{ "prompt": "vivid artistic style, women wearing a pinstripe suit", "structure": "tile", "input_image": "https://replicate.delivery/pbxt/L8R5MODm9TDPWZnUMbAR06XQffs4NuBCJt0PML42PDd8s2vy/tile.jpg", "output_format": "webp", "control_weight": 0.3, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" }
Install Replicateβs 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 zsxkib/sd3-controlnet using Replicateβs API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zsxkib/sd3-controlnet:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897", { input: { prompt: "vivid artistic style, women wearing a pinstripe suit", structure: "tile", input_image: "https://replicate.delivery/pbxt/L8R5MODm9TDPWZnUMbAR06XQffs4NuBCJt0PML42PDd8s2vy/tile.jpg", output_format: "webp", control_weight: 0.3, guidance_scale: 7, output_quality: 80, inference_steps: 25, negative_prompt: "NSFW, nude, naked, porn, ugly" } } ); // To access the file URL: console.log(output.url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicateβs Python client library:pip install replicate
Import the client:import replicate
Run zsxkib/sd3-controlnet using Replicateβs API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zsxkib/sd3-controlnet:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897", input={ "prompt": "vivid artistic style, women wearing a pinstripe suit", "structure": "tile", "input_image": "https://replicate.delivery/pbxt/L8R5MODm9TDPWZnUMbAR06XQffs4NuBCJt0PML42PDd8s2vy/tile.jpg", "output_format": "webp", "control_weight": 0.3, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" } ) # To access the file URL: print(output.url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output.read())
To learn more, take a look at the guide on getting started with Python.
Run zsxkib/sd3-controlnet using Replicateβs 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": "zsxkib/sd3-controlnet:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897", "input": { "prompt": "vivid artistic style, women wearing a pinstripe suit", "structure": "tile", "input_image": "https://replicate.delivery/pbxt/L8R5MODm9TDPWZnUMbAR06XQffs4NuBCJt0PML42PDd8s2vy/tile.jpg", "output_format": "webp", "control_weight": 0.3, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicateβs HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you donβt 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/zsxkib/sd3-controlnet@sha256:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897 \ -i 'prompt="vivid artistic style, women wearing a pinstripe suit"' \ -i 'structure="tile"' \ -i 'input_image="https://replicate.delivery/pbxt/L8R5MODm9TDPWZnUMbAR06XQffs4NuBCJt0PML42PDd8s2vy/tile.jpg"' \ -i 'output_format="webp"' \ -i 'control_weight=0.3' \ -i 'guidance_scale=7' \ -i 'output_quality=80' \ -i 'inference_steps=25' \ -i 'negative_prompt="NSFW, nude, naked, porn, ugly"'
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/zsxkib/sd3-controlnet@sha256:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "prompt": "vivid artistic style, women wearing a pinstripe suit", "structure": "tile", "input_image": "https://replicate.delivery/pbxt/L8R5MODm9TDPWZnUMbAR06XQffs4NuBCJt0PML42PDd8s2vy/tile.jpg", "output_format": "webp", "control_weight": 0.3, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2024-06-21T12:42:49.189217Z", "created_at": "2024-06-21T12:42:23.647000Z", "data_removed": false, "error": null, "id": "r62c05e23xrgp0cg7b8ssxesmr", "input": { "prompt": "vivid artistic style, women wearing a pinstripe suit", "structure": "tile", "input_image": "https://replicate.delivery/pbxt/L8R5MODm9TDPWZnUMbAR06XQffs4NuBCJt0PML42PDd8s2vy/tile.jpg", "output_format": "webp", "control_weight": 0.3, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" }, "logs": "Using seed: 42020\nLoading pipeline components...: 0%| | 0/9 [00:00<?, ?it/s]\nLoading pipeline components...: 11%|β | 1/9 [00:00<00:01, 4.30it/s]\nLoading pipeline components...: 44%|βββββ | 4/9 [00:03<00:05, 1.03s/it]\nLoading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s]\u001b[A\nLoading checkpoint shards: 50%|βββββ | 1/2 [00:01<00:01, 1.27s/it]\u001b[A\nLoading checkpoint shards: 100%|ββββββββββ| 2/2 [00:02<00:00, 1.05s/it]\u001b[A\nLoading checkpoint shards: 100%|ββββββββββ| 2/2 [00:02<00:00, 1.08s/it]\nLoading pipeline components...: 56%|ββββββ | 5/9 [00:06<00:05, 1.36s/it]\nLoading pipeline components...: 78%|ββββββββ | 7/9 [00:06<00:01, 1.25it/s]\nLoading pipeline components...: 100%|ββββββββββ| 9/9 [00:06<00:00, 1.67it/s]\nLoading pipeline components...: 100%|ββββββββββ| 9/9 [00:06<00:00, 1.32it/s]\n 0%| | 0/25 [00:00<?, ?it/s]\n 4%|β | 1/25 [00:00<00:07, 3.32it/s]\n 8%|β | 2/25 [00:00<00:06, 3.32it/s]\n 12%|ββ | 3/25 [00:00<00:06, 3.30it/s]\n 16%|ββ | 4/25 [00:01<00:06, 3.30it/s]\n 20%|ββ | 5/25 [00:01<00:06, 3.31it/s]\n 24%|βββ | 6/25 [00:01<00:05, 3.30it/s]\n 28%|βββ | 7/25 [00:02<00:05, 3.30it/s]\n 32%|ββββ | 8/25 [00:02<00:05, 3.30it/s]\n 36%|ββββ | 9/25 [00:02<00:04, 3.30it/s]\n 40%|ββββ | 10/25 [00:03<00:04, 3.29it/s]\n 44%|βββββ | 11/25 [00:03<00:04, 3.30it/s]\n 48%|βββββ | 12/25 [00:03<00:03, 3.30it/s]\n 52%|ββββββ | 13/25 [00:03<00:03, 3.29it/s]\n 56%|ββββββ | 14/25 [00:04<00:03, 3.29it/s]\n 60%|ββββββ | 15/25 [00:04<00:03, 3.29it/s]\n 64%|βββββββ | 16/25 [00:04<00:02, 3.28it/s]\n 68%|βββββββ | 17/25 [00:05<00:02, 3.28it/s]\n 72%|ββββββββ | 18/25 [00:05<00:02, 3.28it/s]\n 76%|ββββββββ | 19/25 [00:05<00:01, 3.28it/s]\n 80%|ββββββββ | 20/25 [00:06<00:01, 3.28it/s]\n 84%|βββββββββ | 21/25 [00:06<00:01, 3.28it/s]\n 88%|βββββββββ | 22/25 [00:06<00:00, 3.28it/s]\n 92%|ββββββββββ| 23/25 [00:06<00:00, 3.28it/s]\n 96%|ββββββββββ| 24/25 [00:07<00:00, 3.28it/s]\n100%|ββββββββββ| 25/25 [00:07<00:00, 3.28it/s]\n100%|ββββββββββ| 25/25 [00:07<00:00, 3.29it/s]\n[~] Saving to output.webp...\n[~] Output format: WEBP\n[~] Output quality: 80", "metrics": { "predict_time": 23.804016421, "total_time": 25.542217 }, "output": "https://replicate.delivery/pbxt/biya6fiLfDoMEE3ZMwOU3Kf38hnswmx29ZoDdBeHc0wg0JDMB/output.webp", "started_at": "2024-06-21T12:42:25.385200Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/r62c05e23xrgp0cg7b8ssxesmr", "cancel": "https://api.replicate.com/v1/predictions/r62c05e23xrgp0cg7b8ssxesmr/cancel" }, "version": "a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897" }
Generated inUsing seed: 42020 Loading pipeline components...: 0%| | 0/9 [00:00<?, ?it/s] Loading pipeline components...: 11%|β | 1/9 [00:00<00:01, 4.30it/s] Loading pipeline components...: 44%|βββββ | 4/9 [00:03<00:05, 1.03s/it] Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|βββββ | 1/2 [00:01<00:01, 1.27s/it] Loading checkpoint shards: 100%|ββββββββββ| 2/2 [00:02<00:00, 1.05s/it] Loading checkpoint shards: 100%|ββββββββββ| 2/2 [00:02<00:00, 1.08s/it] Loading pipeline components...: 56%|ββββββ | 5/9 [00:06<00:05, 1.36s/it] Loading pipeline components...: 78%|ββββββββ | 7/9 [00:06<00:01, 1.25it/s] Loading pipeline components...: 100%|ββββββββββ| 9/9 [00:06<00:00, 1.67it/s] Loading pipeline components...: 100%|ββββββββββ| 9/9 [00:06<00:00, 1.32it/s] 0%| | 0/25 [00:00<?, ?it/s] 4%|β | 1/25 [00:00<00:07, 3.32it/s] 8%|β | 2/25 [00:00<00:06, 3.32it/s] 12%|ββ | 3/25 [00:00<00:06, 3.30it/s] 16%|ββ | 4/25 [00:01<00:06, 3.30it/s] 20%|ββ | 5/25 [00:01<00:06, 3.31it/s] 24%|βββ | 6/25 [00:01<00:05, 3.30it/s] 28%|βββ | 7/25 [00:02<00:05, 3.30it/s] 32%|ββββ | 8/25 [00:02<00:05, 3.30it/s] 36%|ββββ | 9/25 [00:02<00:04, 3.30it/s] 40%|ββββ | 10/25 [00:03<00:04, 3.29it/s] 44%|βββββ | 11/25 [00:03<00:04, 3.30it/s] 48%|βββββ | 12/25 [00:03<00:03, 3.30it/s] 52%|ββββββ | 13/25 [00:03<00:03, 3.29it/s] 56%|ββββββ | 14/25 [00:04<00:03, 3.29it/s] 60%|ββββββ | 15/25 [00:04<00:03, 3.29it/s] 64%|βββββββ | 16/25 [00:04<00:02, 3.28it/s] 68%|βββββββ | 17/25 [00:05<00:02, 3.28it/s] 72%|ββββββββ | 18/25 [00:05<00:02, 3.28it/s] 76%|ββββββββ | 19/25 [00:05<00:01, 3.28it/s] 80%|ββββββββ | 20/25 [00:06<00:01, 3.28it/s] 84%|βββββββββ | 21/25 [00:06<00:01, 3.28it/s] 88%|βββββββββ | 22/25 [00:06<00:00, 3.28it/s] 92%|ββββββββββ| 23/25 [00:06<00:00, 3.28it/s] 96%|ββββββββββ| 24/25 [00:07<00:00, 3.28it/s] 100%|ββββββββββ| 25/25 [00:07<00:00, 3.28it/s] 100%|ββββββββββ| 25/25 [00:07<00:00, 3.29it/s] [~] Saving to output.webp... [~] Output format: WEBP [~] Output quality: 80
Prediction
zsxkib/sd3-controlnet:3fba667a81db1bdecbe641ed1a9d2ad04d532a3b543ce5d44acc96bf1fe7b733IDzgrk8wzptnrgg0cg672s9x49drStatusSucceededSourceWebHardwareA40 (Large)Total durationCreatedby @zsxkibInput
- prompt
- A detailed image of a hand holding a smartphone. The smartphone is made entirely of vivid, colorful ecclesiastical stained glass, with intricate designs typical of church windows. The stained glass elements should be translucent, allowing light to pass through and creating a vibrant, luminous effect. The hand should be realistic and gently cradling the smartphone, emphasizing the contrast between modern technology and traditional stained glass artistry. Background should be neutral to highlight the details of the stained glass smartphone.
- output_format
- webp
- control_weight
- 0.9
- guidance_scale
- 7
- output_quality
- 80
- inference_steps
- 25
- negative_prompt
- NSFW, nude, naked, porn, ugly
{ "prompt": "A detailed image of a hand holding a smartphone. The smartphone is made entirely of vivid, colorful ecclesiastical stained glass, with intricate designs typical of church windows. The stained glass elements should be translucent, allowing light to pass through and creating a vibrant, luminous effect. The hand should be realistic and gently cradling the smartphone, emphasizing the contrast between modern technology and traditional stained glass artistry. Background should be neutral to highlight the details of the stained glass smartphone.", "input_image": "https://replicate.delivery/pbxt/L7oQYWUw0SvS09BV8hSoV6wAB31oKPn4w9V0LP06Q2F98jnJ/Hand-Holding-Smartphone-Mockup-2-1536x1136.jpg", "output_format": "webp", "control_weight": 0.9, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" }
Install Replicateβs 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 zsxkib/sd3-controlnet using Replicateβs API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zsxkib/sd3-controlnet:3fba667a81db1bdecbe641ed1a9d2ad04d532a3b543ce5d44acc96bf1fe7b733", { input: { prompt: "A detailed image of a hand holding a smartphone. The smartphone is made entirely of vivid, colorful ecclesiastical stained glass, with intricate designs typical of church windows. The stained glass elements should be translucent, allowing light to pass through and creating a vibrant, luminous effect. The hand should be realistic and gently cradling the smartphone, emphasizing the contrast between modern technology and traditional stained glass artistry. Background should be neutral to highlight the details of the stained glass smartphone.", input_image: "https://replicate.delivery/pbxt/L7oQYWUw0SvS09BV8hSoV6wAB31oKPn4w9V0LP06Q2F98jnJ/Hand-Holding-Smartphone-Mockup-2-1536x1136.jpg", output_format: "webp", control_weight: 0.9, guidance_scale: 7, output_quality: 80, inference_steps: 25, negative_prompt: "NSFW, nude, naked, porn, ugly" } } ); // To access the file URL: console.log(output.url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicateβs Python client library:pip install replicate
Import the client:import replicate
Run zsxkib/sd3-controlnet using Replicateβs API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zsxkib/sd3-controlnet:3fba667a81db1bdecbe641ed1a9d2ad04d532a3b543ce5d44acc96bf1fe7b733", input={ "prompt": "A detailed image of a hand holding a smartphone. The smartphone is made entirely of vivid, colorful ecclesiastical stained glass, with intricate designs typical of church windows. The stained glass elements should be translucent, allowing light to pass through and creating a vibrant, luminous effect. The hand should be realistic and gently cradling the smartphone, emphasizing the contrast between modern technology and traditional stained glass artistry. Background should be neutral to highlight the details of the stained glass smartphone.", "input_image": "https://replicate.delivery/pbxt/L7oQYWUw0SvS09BV8hSoV6wAB31oKPn4w9V0LP06Q2F98jnJ/Hand-Holding-Smartphone-Mockup-2-1536x1136.jpg", "output_format": "webp", "control_weight": 0.9, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" } ) # To access the file URL: print(output.url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output.read())
To learn more, take a look at the guide on getting started with Python.
Run zsxkib/sd3-controlnet using Replicateβs 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": "zsxkib/sd3-controlnet:3fba667a81db1bdecbe641ed1a9d2ad04d532a3b543ce5d44acc96bf1fe7b733", "input": { "prompt": "A detailed image of a hand holding a smartphone. The smartphone is made entirely of vivid, colorful ecclesiastical stained glass, with intricate designs typical of church windows. The stained glass elements should be translucent, allowing light to pass through and creating a vibrant, luminous effect. The hand should be realistic and gently cradling the smartphone, emphasizing the contrast between modern technology and traditional stained glass artistry. Background should be neutral to highlight the details of the stained glass smartphone.", "input_image": "https://replicate.delivery/pbxt/L7oQYWUw0SvS09BV8hSoV6wAB31oKPn4w9V0LP06Q2F98jnJ/Hand-Holding-Smartphone-Mockup-2-1536x1136.jpg", "output_format": "webp", "control_weight": 0.9, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicateβs HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you donβt 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/zsxkib/sd3-controlnet@sha256:3fba667a81db1bdecbe641ed1a9d2ad04d532a3b543ce5d44acc96bf1fe7b733 \ -i 'prompt="A detailed image of a hand holding a smartphone. The smartphone is made entirely of vivid, colorful ecclesiastical stained glass, with intricate designs typical of church windows. The stained glass elements should be translucent, allowing light to pass through and creating a vibrant, luminous effect. The hand should be realistic and gently cradling the smartphone, emphasizing the contrast between modern technology and traditional stained glass artistry. Background should be neutral to highlight the details of the stained glass smartphone."' \ -i 'input_image="https://replicate.delivery/pbxt/L7oQYWUw0SvS09BV8hSoV6wAB31oKPn4w9V0LP06Q2F98jnJ/Hand-Holding-Smartphone-Mockup-2-1536x1136.jpg"' \ -i 'output_format="webp"' \ -i 'control_weight=0.9' \ -i 'guidance_scale=7' \ -i 'output_quality=80' \ -i 'inference_steps=25' \ -i 'negative_prompt="NSFW, nude, naked, porn, ugly"'
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/zsxkib/sd3-controlnet@sha256:3fba667a81db1bdecbe641ed1a9d2ad04d532a3b543ce5d44acc96bf1fe7b733
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "prompt": "A detailed image of a hand holding a smartphone. The smartphone is made entirely of vivid, colorful ecclesiastical stained glass, with intricate designs typical of church windows. The stained glass elements should be translucent, allowing light to pass through and creating a vibrant, luminous effect. The hand should be realistic and gently cradling the smartphone, emphasizing the contrast between modern technology and traditional stained glass artistry. Background should be neutral to highlight the details of the stained glass smartphone.", "input_image": "https://replicate.delivery/pbxt/L7oQYWUw0SvS09BV8hSoV6wAB31oKPn4w9V0LP06Q2F98jnJ/Hand-Holding-Smartphone-Mockup-2-1536x1136.jpg", "output_format": "webp", "control_weight": 0.9, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2024-06-19T18:33:05.896569Z", "created_at": "2024-06-19T18:32:55.765000Z", "data_removed": false, "error": null, "id": "zgrk8wzptnrgg0cg672s9x49dr", "input": { "prompt": "A detailed image of a hand holding a smartphone. The smartphone is made entirely of vivid, colorful ecclesiastical stained glass, with intricate designs typical of church windows. The stained glass elements should be translucent, allowing light to pass through and creating a vibrant, luminous effect. The hand should be realistic and gently cradling the smartphone, emphasizing the contrast between modern technology and traditional stained glass artistry. Background should be neutral to highlight the details of the stained glass smartphone.", "input_image": "https://replicate.delivery/pbxt/L7oQYWUw0SvS09BV8hSoV6wAB31oKPn4w9V0LP06Q2F98jnJ/Hand-Holding-Smartphone-Mockup-2-1536x1136.jpg", "output_format": "webp", "control_weight": 0.9, "guidance_scale": 7, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly" }, "logs": "Using seed: 34433\nToken indices sequence length is longer than the specified maximum sequence length for this model (96 > 77). Running this sequence through the model will result in indexing errors\nThe following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['traditional stained glass artistry. background should be neutral to highlight the details of the stained glass smartphone.']\nToken indices sequence length is longer than the specified maximum sequence length for this model (96 > 77). Running this sequence through the model will result in indexing errors\nThe following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['traditional stained glass artistry. background should be neutral to highlight the details of the stained glass smartphone.']\n 0%| | 0/25 [00:00<?, ?it/s]\n 4%|β | 1/25 [00:00<00:07, 3.31it/s]\n 8%|β | 2/25 [00:00<00:06, 3.30it/s]\n 12%|ββ | 3/25 [00:00<00:06, 3.26it/s]\n 16%|ββ | 4/25 [00:01<00:06, 3.27it/s]\n 20%|ββ | 5/25 [00:01<00:06, 3.27it/s]\n 24%|βββ | 6/25 [00:01<00:05, 3.26it/s]\n 28%|βββ | 7/25 [00:02<00:05, 3.26it/s]\n 32%|ββββ | 8/25 [00:02<00:05, 3.26it/s]\n 36%|ββββ | 9/25 [00:02<00:04, 3.26it/s]\n 40%|ββββ | 10/25 [00:03<00:04, 3.25it/s]\n 44%|βββββ | 11/25 [00:03<00:04, 3.26it/s]\n 48%|βββββ | 12/25 [00:03<00:03, 3.25it/s]\n 52%|ββββββ | 13/25 [00:03<00:03, 3.24it/s]\n 56%|ββββββ | 14/25 [00:04<00:03, 3.25it/s]\n 60%|ββββββ | 15/25 [00:04<00:03, 3.26it/s]\n 64%|βββββββ | 16/25 [00:04<00:02, 3.26it/s]\n 68%|βββββββ | 17/25 [00:05<00:02, 3.27it/s]\n 72%|ββββββββ | 18/25 [00:05<00:02, 3.28it/s]\n 76%|ββββββββ | 19/25 [00:05<00:01, 3.27it/s]\n 80%|ββββββββ | 20/25 [00:06<00:01, 3.28it/s]\n 84%|βββββββββ | 21/25 [00:06<00:01, 3.28it/s]\n 88%|βββββββββ | 22/25 [00:06<00:00, 3.28it/s]\n 92%|ββββββββββ| 23/25 [00:07<00:00, 3.28it/s]\n 96%|ββββββββββ| 24/25 [00:07<00:00, 3.28it/s]\n100%|ββββββββββ| 25/25 [00:07<00:00, 3.28it/s]\n100%|ββββββββββ| 25/25 [00:07<00:00, 3.27it/s]\n[~] Saving to output.webp...\n[~] Output format: WEBP\n[~] Output quality: 80", "metrics": { "predict_time": 10.088725134, "total_time": 10.131569 }, "output": "https://replicate.delivery/pbxt/XJWcbqC9ylInEddwzVODUZJM8QRg0DdXWOjeWECw2o4wsGgJA/output.webp", "started_at": "2024-06-19T18:32:55.807844Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/zgrk8wzptnrgg0cg672s9x49dr", "cancel": "https://api.replicate.com/v1/predictions/zgrk8wzptnrgg0cg672s9x49dr/cancel" }, "version": "3fba667a81db1bdecbe641ed1a9d2ad04d532a3b543ce5d44acc96bf1fe7b733" }
Generated inUsing seed: 34433 Token indices sequence length is longer than the specified maximum sequence length for this model (96 > 77). Running this sequence through the model will result in indexing errors The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['traditional stained glass artistry. background should be neutral to highlight the details of the stained glass smartphone.'] Token indices sequence length is longer than the specified maximum sequence length for this model (96 > 77). Running this sequence through the model will result in indexing errors The following part of your input was truncated because CLIP can only handle sequences up to 77 tokens: ['traditional stained glass artistry. background should be neutral to highlight the details of the stained glass smartphone.'] 0%| | 0/25 [00:00<?, ?it/s] 4%|β | 1/25 [00:00<00:07, 3.31it/s] 8%|β | 2/25 [00:00<00:06, 3.30it/s] 12%|ββ | 3/25 [00:00<00:06, 3.26it/s] 16%|ββ | 4/25 [00:01<00:06, 3.27it/s] 20%|ββ | 5/25 [00:01<00:06, 3.27it/s] 24%|βββ | 6/25 [00:01<00:05, 3.26it/s] 28%|βββ | 7/25 [00:02<00:05, 3.26it/s] 32%|ββββ | 8/25 [00:02<00:05, 3.26it/s] 36%|ββββ | 9/25 [00:02<00:04, 3.26it/s] 40%|ββββ | 10/25 [00:03<00:04, 3.25it/s] 44%|βββββ | 11/25 [00:03<00:04, 3.26it/s] 48%|βββββ | 12/25 [00:03<00:03, 3.25it/s] 52%|ββββββ | 13/25 [00:03<00:03, 3.24it/s] 56%|ββββββ | 14/25 [00:04<00:03, 3.25it/s] 60%|ββββββ | 15/25 [00:04<00:03, 3.26it/s] 64%|βββββββ | 16/25 [00:04<00:02, 3.26it/s] 68%|βββββββ | 17/25 [00:05<00:02, 3.27it/s] 72%|ββββββββ | 18/25 [00:05<00:02, 3.28it/s] 76%|ββββββββ | 19/25 [00:05<00:01, 3.27it/s] 80%|ββββββββ | 20/25 [00:06<00:01, 3.28it/s] 84%|βββββββββ | 21/25 [00:06<00:01, 3.28it/s] 88%|βββββββββ | 22/25 [00:06<00:00, 3.28it/s] 92%|ββββββββββ| 23/25 [00:07<00:00, 3.28it/s] 96%|ββββββββββ| 24/25 [00:07<00:00, 3.28it/s] 100%|ββββββββββ| 25/25 [00:07<00:00, 3.28it/s] 100%|ββββββββββ| 25/25 [00:07<00:00, 3.27it/s] [~] Saving to output.webp... [~] Output format: WEBP [~] Output quality: 80
Prediction
zsxkib/sd3-controlnet:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897ID4fk5h3cpbdrgj0cg7bdt5gxym4StatusSucceededSourceWebHardwareA40 (Large)Total durationCreatedInput
- prompt
- A stunning realistic dynamic shot of an astronaut with cowboy hat
- structure
- pose
- output_format
- webp
- control_weight
- 0.5
- guidance_scale
- 6
- output_quality
- 80
- inference_steps
- 25
- negative_prompt
- NSFW, nude, naked, porn, ugly, helmet, astronaut helmet
{ "prompt": "A stunning realistic dynamic shot of an astronaut with cowboy hat", "structure": "pose", "input_image": "https://replicate.delivery/pbxt/L8R4ydzoL5fz4qIDSHi0R4PllEaGyV3pMjQyuNZQubhotI74/face-pose.jpg", "output_format": "webp", "control_weight": 0.5, "guidance_scale": 6, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly, helmet, astronaut helmet" }
Install Replicateβs 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 zsxkib/sd3-controlnet using Replicateβs API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zsxkib/sd3-controlnet:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897", { input: { prompt: "A stunning realistic dynamic shot of an astronaut with cowboy hat", structure: "pose", input_image: "https://replicate.delivery/pbxt/L8R4ydzoL5fz4qIDSHi0R4PllEaGyV3pMjQyuNZQubhotI74/face-pose.jpg", output_format: "webp", control_weight: 0.5, guidance_scale: 6, output_quality: 80, inference_steps: 25, negative_prompt: "NSFW, nude, naked, porn, ugly, helmet, astronaut helmet" } } ); // To access the file URL: console.log(output.url()); //=> "http://example.com" // To write the file to disk: fs.writeFile("my-image.png", output);
To learn more, take a look at the guide on getting started with Node.js.
Install Replicateβs Python client library:pip install replicate
Import the client:import replicate
Run zsxkib/sd3-controlnet using Replicateβs API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zsxkib/sd3-controlnet:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897", input={ "prompt": "A stunning realistic dynamic shot of an astronaut with cowboy hat", "structure": "pose", "input_image": "https://replicate.delivery/pbxt/L8R4ydzoL5fz4qIDSHi0R4PllEaGyV3pMjQyuNZQubhotI74/face-pose.jpg", "output_format": "webp", "control_weight": 0.5, "guidance_scale": 6, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly, helmet, astronaut helmet" } ) # To access the file URL: print(output.url()) #=> "http://example.com" # To write the file to disk: with open("my-image.png", "wb") as file: file.write(output.read())
To learn more, take a look at the guide on getting started with Python.
Run zsxkib/sd3-controlnet using Replicateβs 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": "zsxkib/sd3-controlnet:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897", "input": { "prompt": "A stunning realistic dynamic shot of an astronaut with cowboy hat", "structure": "pose", "input_image": "https://replicate.delivery/pbxt/L8R4ydzoL5fz4qIDSHi0R4PllEaGyV3pMjQyuNZQubhotI74/face-pose.jpg", "output_format": "webp", "control_weight": 0.5, "guidance_scale": 6, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly, helmet, astronaut helmet" } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicateβs HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you donβt 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/zsxkib/sd3-controlnet@sha256:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897 \ -i 'prompt="A stunning realistic dynamic shot of an astronaut with cowboy hat"' \ -i 'structure="pose"' \ -i 'input_image="https://replicate.delivery/pbxt/L8R4ydzoL5fz4qIDSHi0R4PllEaGyV3pMjQyuNZQubhotI74/face-pose.jpg"' \ -i 'output_format="webp"' \ -i 'control_weight=0.5' \ -i 'guidance_scale=6' \ -i 'output_quality=80' \ -i 'inference_steps=25' \ -i 'negative_prompt="NSFW, nude, naked, porn, ugly, helmet, astronaut helmet"'
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/zsxkib/sd3-controlnet@sha256:a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "prompt": "A stunning realistic dynamic shot of an astronaut with cowboy hat", "structure": "pose", "input_image": "https://replicate.delivery/pbxt/L8R4ydzoL5fz4qIDSHi0R4PllEaGyV3pMjQyuNZQubhotI74/face-pose.jpg", "output_format": "webp", "control_weight": 0.5, "guidance_scale": 6, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly, helmet, astronaut helmet" } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2024-06-21T12:53:28.510148Z", "created_at": "2024-06-21T12:53:07.803000Z", "data_removed": false, "error": null, "id": "4fk5h3cpbdrgj0cg7bdt5gxym4", "input": { "prompt": "A stunning realistic dynamic shot of an astronaut with cowboy hat", "structure": "pose", "input_image": "https://replicate.delivery/pbxt/L8R4ydzoL5fz4qIDSHi0R4PllEaGyV3pMjQyuNZQubhotI74/face-pose.jpg", "output_format": "webp", "control_weight": 0.5, "guidance_scale": 6, "output_quality": 80, "inference_steps": 25, "negative_prompt": "NSFW, nude, naked, porn, ugly, helmet, astronaut helmet" }, "logs": "Using seed: 16287\nLoading pipeline components...: 0%| | 0/9 [00:00<?, ?it/s]\nLoading pipeline components...: 11%|β | 1/9 [00:00<00:01, 7.50it/s]\nLoading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s]\u001b[A\nLoading checkpoint shards: 50%|βββββ | 1/2 [00:00<00:00, 1.05it/s]\u001b[A\nLoading checkpoint shards: 100%|ββββββββββ| 2/2 [00:01<00:00, 1.06it/s]\u001b[A\nLoading checkpoint shards: 100%|ββββββββββ| 2/2 [00:01<00:00, 1.06it/s]\nLoading pipeline components...: 33%|ββββ | 3/9 [00:02<00:04, 1.29it/s]\nLoading pipeline components...: 56%|ββββββ | 5/9 [00:02<00:02, 1.95it/s]\nLoading pipeline components...: 78%|ββββββββ | 7/9 [00:04<00:01, 1.45it/s]\nLoading pipeline components...: 89%|βββββββββ | 8/9 [00:04<00:00, 1.74it/s]\nLoading pipeline components...: 100%|ββββββββββ| 9/9 [00:04<00:00, 1.89it/s]\n 0%| | 0/25 [00:00<?, ?it/s]\n 4%|β | 1/25 [00:00<00:07, 3.27it/s]\n 8%|β | 2/25 [00:00<00:07, 3.26it/s]\n 12%|ββ | 3/25 [00:00<00:06, 3.23it/s]\n 16%|ββ | 4/25 [00:01<00:06, 3.24it/s]\n 20%|ββ | 5/25 [00:01<00:06, 3.24it/s]\n 24%|βββ | 6/25 [00:01<00:05, 3.23it/s]\n 28%|βββ | 7/25 [00:02<00:05, 3.23it/s]\n 32%|ββββ | 8/25 [00:02<00:05, 3.24it/s]\n 36%|ββββ | 9/25 [00:02<00:04, 3.23it/s]\n 40%|ββββ | 10/25 [00:03<00:04, 3.24it/s]\n 44%|βββββ | 11/25 [00:03<00:04, 3.25it/s]\n 48%|βββββ | 12/25 [00:03<00:03, 3.26it/s]\n 52%|ββββββ | 13/25 [00:04<00:03, 3.25it/s]\n 56%|ββββββ | 14/25 [00:04<00:03, 3.26it/s]\n 60%|ββββββ | 15/25 [00:04<00:03, 3.26it/s]\n 64%|βββββββ | 16/25 [00:04<00:02, 3.26it/s]\n 68%|βββββββ | 17/25 [00:05<00:02, 3.26it/s]\n 72%|ββββββββ | 18/25 [00:05<00:02, 3.26it/s]\n 76%|ββββββββ | 19/25 [00:05<00:01, 3.26it/s]\n 80%|ββββββββ | 20/25 [00:06<00:01, 3.26it/s]\n 84%|βββββββββ | 21/25 [00:06<00:01, 3.26it/s]\n 88%|βββββββββ | 22/25 [00:06<00:00, 3.26it/s]\n 92%|ββββββββββ| 23/25 [00:07<00:00, 3.25it/s]\n 96%|ββββββββββ| 24/25 [00:07<00:00, 3.26it/s]\n100%|ββββββββββ| 25/25 [00:07<00:00, 3.25it/s]\n100%|ββββββββββ| 25/25 [00:07<00:00, 3.25it/s]\n[~] Saving to output.webp...\n[~] Output format: WEBP\n[~] Output quality: 80", "metrics": { "predict_time": 20.667831938, "total_time": 20.707148 }, "output": "https://replicate.delivery/pbxt/ye4hfwfFtudVSIcGL2eB9nScPwe56FQTiAfeyPDv819jjTZgJA/output.webp", "started_at": "2024-06-21T12:53:07.842316Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/4fk5h3cpbdrgj0cg7bdt5gxym4", "cancel": "https://api.replicate.com/v1/predictions/4fk5h3cpbdrgj0cg7bdt5gxym4/cancel" }, "version": "a8d23dad5ee7c2ac7269723fa8c1480da9b6edbfec17f6291f9d6c88835d9897" }
Generated inUsing seed: 16287 Loading pipeline components...: 0%| | 0/9 [00:00<?, ?it/s] Loading pipeline components...: 11%|β | 1/9 [00:00<00:01, 7.50it/s] Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|βββββ | 1/2 [00:00<00:00, 1.05it/s] Loading checkpoint shards: 100%|ββββββββββ| 2/2 [00:01<00:00, 1.06it/s] Loading checkpoint shards: 100%|ββββββββββ| 2/2 [00:01<00:00, 1.06it/s] Loading pipeline components...: 33%|ββββ | 3/9 [00:02<00:04, 1.29it/s] Loading pipeline components...: 56%|ββββββ | 5/9 [00:02<00:02, 1.95it/s] Loading pipeline components...: 78%|ββββββββ | 7/9 [00:04<00:01, 1.45it/s] Loading pipeline components...: 89%|βββββββββ | 8/9 [00:04<00:00, 1.74it/s] Loading pipeline components...: 100%|ββββββββββ| 9/9 [00:04<00:00, 1.89it/s] 0%| | 0/25 [00:00<?, ?it/s] 4%|β | 1/25 [00:00<00:07, 3.27it/s] 8%|β | 2/25 [00:00<00:07, 3.26it/s] 12%|ββ | 3/25 [00:00<00:06, 3.23it/s] 16%|ββ | 4/25 [00:01<00:06, 3.24it/s] 20%|ββ | 5/25 [00:01<00:06, 3.24it/s] 24%|βββ | 6/25 [00:01<00:05, 3.23it/s] 28%|βββ | 7/25 [00:02<00:05, 3.23it/s] 32%|ββββ | 8/25 [00:02<00:05, 3.24it/s] 36%|ββββ | 9/25 [00:02<00:04, 3.23it/s] 40%|ββββ | 10/25 [00:03<00:04, 3.24it/s] 44%|βββββ | 11/25 [00:03<00:04, 3.25it/s] 48%|βββββ | 12/25 [00:03<00:03, 3.26it/s] 52%|ββββββ | 13/25 [00:04<00:03, 3.25it/s] 56%|ββββββ | 14/25 [00:04<00:03, 3.26it/s] 60%|ββββββ | 15/25 [00:04<00:03, 3.26it/s] 64%|βββββββ | 16/25 [00:04<00:02, 3.26it/s] 68%|βββββββ | 17/25 [00:05<00:02, 3.26it/s] 72%|ββββββββ | 18/25 [00:05<00:02, 3.26it/s] 76%|ββββββββ | 19/25 [00:05<00:01, 3.26it/s] 80%|ββββββββ | 20/25 [00:06<00:01, 3.26it/s] 84%|βββββββββ | 21/25 [00:06<00:01, 3.26it/s] 88%|βββββββββ | 22/25 [00:06<00:00, 3.26it/s] 92%|ββββββββββ| 23/25 [00:07<00:00, 3.25it/s] 96%|ββββββββββ| 24/25 [00:07<00:00, 3.26it/s] 100%|ββββββββββ| 25/25 [00:07<00:00, 3.25it/s] 100%|ββββββββββ| 25/25 [00:07<00:00, 3.25it/s] [~] Saving to output.webp... [~] Output format: WEBP [~] Output quality: 80
Prediction
zsxkib/sd3-controlnet:1512680a528ba127ad49e92e36780974e5aa12c7d717af83c4a98a9253adf2c6ID9g6b0dqftdrgg0cg7h49p5tx44StatusSucceededSourceWebHardwareA40 (Large)Total durationCreatedInput
- prompt
- A surreal interpretation of an ice cream cone transforming into a fantastical cloud city, with miniature buildings and flying vehicles emerging from the swirls of ice cream
- structure
- canny
- num_outputs
- 1
- aspect_ratio
- 1:1
- low_threshold
- 1
- output_format
- webp
- control_weight
- 0.7
- guidance_scale
- 7
- high_threshold
- 255
- output_quality
- 90
- inference_steps
- 20
- negative_prompt
- NSFW, nude, naked, porn, ugly, melting, dripping, realistic ice cream
{ "prompt": "A surreal interpretation of an ice cream cone transforming into a fantastical cloud city, with miniature buildings and flying vehicles emerging from the swirls of ice cream", "structure": "canny", "input_image": "https://replicate.delivery/pbxt/L8XLfd3JY13a51BlD1xA1APD4lPCqFgVJD5pG00Xpr4klsbQ/canny.png", "num_outputs": 1, "aspect_ratio": "1:1", "low_threshold": 1, "output_format": "webp", "control_weight": 0.7, "guidance_scale": 7, "high_threshold": 255, "output_quality": 90, "inference_steps": 20, "negative_prompt": "NSFW, nude, naked, porn, ugly, melting, dripping, realistic ice cream" }
Install Replicateβs 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 zsxkib/sd3-controlnet using Replicateβs API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zsxkib/sd3-controlnet:1512680a528ba127ad49e92e36780974e5aa12c7d717af83c4a98a9253adf2c6", { input: { prompt: "A surreal interpretation of an ice cream cone transforming into a fantastical cloud city, with miniature buildings and flying vehicles emerging from the swirls of ice cream", structure: "canny", input_image: "https://replicate.delivery/pbxt/L8XLfd3JY13a51BlD1xA1APD4lPCqFgVJD5pG00Xpr4klsbQ/canny.png", num_outputs: 1, aspect_ratio: "1:1", low_threshold: 1, output_format: "webp", control_weight: 0.7, guidance_scale: 7, high_threshold: 255, output_quality: 90, inference_steps: 20, negative_prompt: "NSFW, nude, naked, porn, ugly, melting, dripping, realistic ice cream" } } ); // 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 Replicateβs Python client library:pip install replicate
Import the client:import replicate
Run zsxkib/sd3-controlnet using Replicateβs API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zsxkib/sd3-controlnet:1512680a528ba127ad49e92e36780974e5aa12c7d717af83c4a98a9253adf2c6", input={ "prompt": "A surreal interpretation of an ice cream cone transforming into a fantastical cloud city, with miniature buildings and flying vehicles emerging from the swirls of ice cream", "structure": "canny", "input_image": "https://replicate.delivery/pbxt/L8XLfd3JY13a51BlD1xA1APD4lPCqFgVJD5pG00Xpr4klsbQ/canny.png", "num_outputs": 1, "aspect_ratio": "1:1", "low_threshold": 1, "output_format": "webp", "control_weight": 0.7, "guidance_scale": 7, "high_threshold": 255, "output_quality": 90, "inference_steps": 20, "negative_prompt": "NSFW, nude, naked, porn, ugly, melting, dripping, realistic ice cream" } ) # 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 zsxkib/sd3-controlnet using Replicateβs 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": "zsxkib/sd3-controlnet:1512680a528ba127ad49e92e36780974e5aa12c7d717af83c4a98a9253adf2c6", "input": { "prompt": "A surreal interpretation of an ice cream cone transforming into a fantastical cloud city, with miniature buildings and flying vehicles emerging from the swirls of ice cream", "structure": "canny", "input_image": "https://replicate.delivery/pbxt/L8XLfd3JY13a51BlD1xA1APD4lPCqFgVJD5pG00Xpr4klsbQ/canny.png", "num_outputs": 1, "aspect_ratio": "1:1", "low_threshold": 1, "output_format": "webp", "control_weight": 0.7, "guidance_scale": 7, "high_threshold": 255, "output_quality": 90, "inference_steps": 20, "negative_prompt": "NSFW, nude, naked, porn, ugly, melting, dripping, realistic ice cream" } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicateβs HTTP API reference docs.
You can run this model locally using Cog. First, install Cog:brew install cog
If you donβt 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/zsxkib/sd3-controlnet@sha256:1512680a528ba127ad49e92e36780974e5aa12c7d717af83c4a98a9253adf2c6 \ -i 'prompt="A surreal interpretation of an ice cream cone transforming into a fantastical cloud city, with miniature buildings and flying vehicles emerging from the swirls of ice cream"' \ -i 'structure="canny"' \ -i 'input_image="https://replicate.delivery/pbxt/L8XLfd3JY13a51BlD1xA1APD4lPCqFgVJD5pG00Xpr4klsbQ/canny.png"' \ -i 'num_outputs=1' \ -i 'aspect_ratio="1:1"' \ -i 'low_threshold=1' \ -i 'output_format="webp"' \ -i 'control_weight=0.7' \ -i 'guidance_scale=7' \ -i 'high_threshold=255' \ -i 'output_quality=90' \ -i 'inference_steps=20' \ -i 'negative_prompt="NSFW, nude, naked, porn, ugly, melting, dripping, realistic ice cream"'
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/zsxkib/sd3-controlnet@sha256:1512680a528ba127ad49e92e36780974e5aa12c7d717af83c4a98a9253adf2c6
curl -s -X POST \ -H "Content-Type: application/json" \ -d $'{ "input": { "prompt": "A surreal interpretation of an ice cream cone transforming into a fantastical cloud city, with miniature buildings and flying vehicles emerging from the swirls of ice cream", "structure": "canny", "input_image": "https://replicate.delivery/pbxt/L8XLfd3JY13a51BlD1xA1APD4lPCqFgVJD5pG00Xpr4klsbQ/canny.png", "num_outputs": 1, "aspect_ratio": "1:1", "low_threshold": 1, "output_format": "webp", "control_weight": 0.7, "guidance_scale": 7, "high_threshold": 255, "output_quality": 90, "inference_steps": 20, "negative_prompt": "NSFW, nude, naked, porn, ugly, melting, dripping, realistic ice cream" } }' \ http://localhost:5000/predictions
To learn more, take a look at the Cog documentation.
Output
{ "completed_at": "2024-06-21T19:32:19.903292Z", "created_at": "2024-06-21T19:32:11.347000Z", "data_removed": false, "error": null, "id": "9g6b0dqftdrgg0cg7h49p5tx44", "input": { "prompt": "A surreal interpretation of an ice cream cone transforming into a fantastical cloud city, with miniature buildings and flying vehicles emerging from the swirls of ice cream", "structure": "canny", "input_image": "https://replicate.delivery/pbxt/L8XLfd3JY13a51BlD1xA1APD4lPCqFgVJD5pG00Xpr4klsbQ/canny.png", "num_outputs": 1, "aspect_ratio": "1:1", "low_threshold": 1, "output_format": "webp", "control_weight": 0.7, "guidance_scale": 7, "high_threshold": 255, "output_quality": 90, "inference_steps": 20, "negative_prompt": "NSFW, nude, naked, porn, ugly, melting, dripping, realistic ice cream" }, "logs": "Using seed: 41082\nReusing existing pipeline for structure: canny\n 0%| | 0/20 [00:00<?, ?it/s]\n 5%|β | 1/20 [00:00<00:05, 3.32it/s]\n 10%|β | 2/20 [00:00<00:05, 3.30it/s]\n 15%|ββ | 3/20 [00:00<00:05, 3.28it/s]\n 20%|ββ | 4/20 [00:01<00:04, 3.29it/s]\n 25%|βββ | 5/20 [00:01<00:04, 3.29it/s]\n 30%|βββ | 6/20 [00:01<00:04, 3.29it/s]\n 35%|ββββ | 7/20 [00:02<00:03, 3.28it/s]\n 40%|ββββ | 8/20 [00:02<00:03, 3.28it/s]\n 45%|βββββ | 9/20 [00:02<00:03, 3.28it/s]\n 50%|βββββ | 10/20 [00:03<00:03, 3.28it/s]\n 55%|ββββββ | 11/20 [00:03<00:02, 3.29it/s]\n 60%|ββββββ | 12/20 [00:03<00:02, 3.30it/s]\n 65%|βββββββ | 13/20 [00:03<00:02, 3.30it/s]\n 70%|βββββββ | 14/20 [00:04<00:01, 3.30it/s]\n 75%|ββββββββ | 15/20 [00:04<00:01, 3.30it/s]\n 80%|ββββββββ | 16/20 [00:04<00:01, 3.31it/s]\n 85%|βββββββββ | 17/20 [00:05<00:00, 3.31it/s]\n 90%|βββββββββ | 18/20 [00:05<00:00, 3.30it/s]\n 95%|ββββββββββ| 19/20 [00:05<00:00, 3.30it/s]\n100%|ββββββββββ| 20/20 [00:06<00:00, 3.30it/s]\n100%|ββββββββββ| 20/20 [00:06<00:00, 3.29it/s]\n[~] Saving to output_0.webp...\n[~] Output format: WEBP\n[~] Output quality: 90", "metrics": { "predict_time": 8.522457477, "total_time": 8.556292 }, "output": [ "https://replicate.delivery/pbxt/nba7X1PHtfXhTCoZ7ZbmnMkjFE4AUeVeQVVenAXZyXjL0hDMB/output_0.webp" ], "started_at": "2024-06-21T19:32:11.380834Z", "status": "succeeded", "urls": { "get": "https://api.replicate.com/v1/predictions/9g6b0dqftdrgg0cg7h49p5tx44", "cancel": "https://api.replicate.com/v1/predictions/9g6b0dqftdrgg0cg7h49p5tx44/cancel" }, "version": "1512680a528ba127ad49e92e36780974e5aa12c7d717af83c4a98a9253adf2c6" }
Generated inUsing seed: 41082 Reusing existing pipeline for structure: canny 0%| | 0/20 [00:00<?, ?it/s] 5%|β | 1/20 [00:00<00:05, 3.32it/s] 10%|β | 2/20 [00:00<00:05, 3.30it/s] 15%|ββ | 3/20 [00:00<00:05, 3.28it/s] 20%|ββ | 4/20 [00:01<00:04, 3.29it/s] 25%|βββ | 5/20 [00:01<00:04, 3.29it/s] 30%|βββ | 6/20 [00:01<00:04, 3.29it/s] 35%|ββββ | 7/20 [00:02<00:03, 3.28it/s] 40%|ββββ | 8/20 [00:02<00:03, 3.28it/s] 45%|βββββ | 9/20 [00:02<00:03, 3.28it/s] 50%|βββββ | 10/20 [00:03<00:03, 3.28it/s] 55%|ββββββ | 11/20 [00:03<00:02, 3.29it/s] 60%|ββββββ | 12/20 [00:03<00:02, 3.30it/s] 65%|βββββββ | 13/20 [00:03<00:02, 3.30it/s] 70%|βββββββ | 14/20 [00:04<00:01, 3.30it/s] 75%|ββββββββ | 15/20 [00:04<00:01, 3.30it/s] 80%|ββββββββ | 16/20 [00:04<00:01, 3.31it/s] 85%|βββββββββ | 17/20 [00:05<00:00, 3.31it/s] 90%|βββββββββ | 18/20 [00:05<00:00, 3.30it/s] 95%|ββββββββββ| 19/20 [00:05<00:00, 3.30it/s] 100%|ββββββββββ| 20/20 [00:06<00:00, 3.30it/s] 100%|ββββββββββ| 20/20 [00:06<00:00, 3.29it/s] [~] Saving to output_0.webp... [~] Output format: WEBP [~] Output quality: 90
Want to make some of these yourself?
Run this model