zurk / hunyuan-video-8bit
Hunyuan Video 8bit model API for video generation (Updated 6 months, 1 week ago)
- Public
- 238 runs
- GitHub
Prediction
zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582bIDpcgh9xbb8hrm80cknc686bkj7rStatusSucceededSourceWebHardwareL40STotal durationCreatedInput
- seed
- 43
- width
- 960
- height
- 544
- prompt
- A cat walks on the grass, realistic style.
- flow_shift
- 7
- video_length
- 65
- negative_prompt
- Ugly
- num_inference_steps
- 40
- embedded_guidance_scale
- 6
{ "seed": 43, "width": 960, "height": 544, "prompt": "A cat walks on the grass, realistic style.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }
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 zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", { input: { seed: 43, width: 960, height: 544, prompt: "A cat walks on the grass, realistic style.", flow_shift: 7, video_length: 65, negative_prompt: "Ugly", num_inference_steps: 40, embedded_guidance_scale: 6 } } ); // 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 Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", input={ "seed": 43, "width": 960, "height": 544, "prompt": "A cat walks on the grass, realistic style.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } ) print(output)
To learn more, take a look at the guide on getting started with Python.
Run zurk/hunyuan-video-8bit 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": "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A cat walks on the grass, realistic style.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
Output
{ "completed_at": "2024-12-09T11:03:14.361471Z", "created_at": "2024-12-09T10:52:18.116000Z", "data_removed": false, "error": null, "id": "pcgh9xbb8hrm80cknc686bkj7r", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A cat walks on the grass, realistic style.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }, "logs": "2024-12-09 10:56:51,578-INFO-cog_predictor-seed=43\n2024-12-09 10:56:59,013-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 65)\n2024-12-09 10:57:01,472-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 65 frames in 17 latents at 960x544 with 40 inference steps\n 0%| | 0/40 [00:00<?, ?it/s]\n 2%|▎ | 1/40 [00:10<06:47, 10.45s/it]\n 5%|▌ | 2/40 [00:19<05:55, 9.36s/it]\n 8%|▊ | 3/40 [00:27<05:34, 9.05s/it]\n 10%|█ | 4/40 [00:36<05:20, 8.90s/it]\n 12%|█▎ | 5/40 [00:45<05:08, 8.82s/it]\n 15%|█▌ | 6/40 [00:53<04:58, 8.79s/it]\n 18%|█▊ | 7/40 [01:02<04:49, 8.77s/it]\n 20%|██ | 8/40 [01:11<04:40, 8.75s/it]\n 22%|██▎ | 9/40 [01:20<04:31, 8.75s/it]\n 25%|██▌ | 10/40 [01:28<04:22, 8.75s/it]\n 28%|██▊ | 11/40 [01:37<04:13, 8.75s/it]\n 30%|███ | 12/40 [01:46<04:04, 8.75s/it]\n 32%|███▎ | 13/40 [01:55<03:56, 8.75s/it]\n 35%|███▌ | 14/40 [02:03<03:47, 8.75s/it]\n 38%|███▊ | 15/40 [02:12<03:38, 8.75s/it]\n 40%|████ | 16/40 [02:21<03:29, 8.75s/it]\n 42%|████▎ | 17/40 [02:29<03:21, 8.75s/it]\n 45%|████▌ | 18/40 [02:38<03:12, 8.75s/it]\n 48%|████▊ | 19/40 [02:47<03:03, 8.75s/it]\n 50%|█████ | 20/40 [02:56<02:54, 8.75s/it]\n 52%|█████▎ | 21/40 [03:04<02:46, 8.75s/it]\n 55%|█████▌ | 22/40 [03:13<02:37, 8.75s/it]\n 57%|█████▊ | 23/40 [03:22<02:28, 8.75s/it]\n 60%|██████ | 24/40 [03:31<02:19, 8.75s/it]\n 62%|██████▎ | 25/40 [03:39<02:11, 8.75s/it]\n 65%|██████▌ | 26/40 [03:48<02:02, 8.75s/it]\n 68%|██████▊ | 27/40 [03:57<01:53, 8.75s/it]\n 70%|███████ | 28/40 [04:06<01:44, 8.75s/it]\n 72%|███████▎ | 29/40 [04:14<01:36, 8.76s/it]\n 75%|███████▌ | 30/40 [04:23<01:27, 8.76s/it]\n 78%|███████▊ | 31/40 [04:32<01:18, 8.76s/it]\n 80%|████████ | 32/40 [04:41<01:10, 8.76s/it]\n 82%|████████▎ | 33/40 [04:50<01:01, 8.76s/it]\n 85%|████████▌ | 34/40 [04:58<00:52, 8.75s/it]\n 88%|████████▊ | 35/40 [05:07<00:43, 8.75s/it]\n 90%|█████████ | 36/40 [05:16<00:35, 8.77s/it]\n 92%|█████████▎| 37/40 [05:25<00:26, 8.76s/it]\n 95%|█████████▌| 38/40 [05:33<00:17, 8.76s/it]\n 98%|█████████▊| 39/40 [05:42<00:08, 8.75s/it]\n100%|██████████| 40/40 [05:51<00:00, 8.76s/it]\n100%|██████████| 40/40 [05:51<00:00, 8.78s/it]\n2024-12-09 11:02:52,817-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.985 GB\n2024-12-09 11:02:52,817-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=17.704 GB\n2024-12-09 11:02:52,817-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=19.467 GB", "metrics": { "predict_time": 382.780903913, "total_time": 656.245471 }, "output": "https://replicate.delivery/xezq/IehvJJjySLyIQyDjuxBjOeQwQeqb9trHMpVfZNAPekTUOAJfE/HunyuanVideo_00008.mp4", "started_at": "2024-12-09T10:56:51.580567Z", "status": "succeeded", "urls": { "stream": "https://stream.replicate.com/v1/files/bcwr-qij7enpot2ct54t2oqifvw6kxn6gxjrulf3zwa6om55ltc5gcczq", "get": "https://api.replicate.com/v1/predictions/pcgh9xbb8hrm80cknc686bkj7r", "cancel": "https://api.replicate.com/v1/predictions/pcgh9xbb8hrm80cknc686bkj7r/cancel" }, "version": "392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b" }
Generated in2024-12-09 10:56:51,578-INFO-cog_predictor-seed=43 2024-12-09 10:56:59,013-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 65) 2024-12-09 10:57:01,472-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 65 frames in 17 latents at 960x544 with 40 inference steps 0%| | 0/40 [00:00<?, ?it/s] 2%|▎ | 1/40 [00:10<06:47, 10.45s/it] 5%|▌ | 2/40 [00:19<05:55, 9.36s/it] 8%|▊ | 3/40 [00:27<05:34, 9.05s/it] 10%|█ | 4/40 [00:36<05:20, 8.90s/it] 12%|█▎ | 5/40 [00:45<05:08, 8.82s/it] 15%|█▌ | 6/40 [00:53<04:58, 8.79s/it] 18%|█▊ | 7/40 [01:02<04:49, 8.77s/it] 20%|██ | 8/40 [01:11<04:40, 8.75s/it] 22%|██▎ | 9/40 [01:20<04:31, 8.75s/it] 25%|██▌ | 10/40 [01:28<04:22, 8.75s/it] 28%|██▊ | 11/40 [01:37<04:13, 8.75s/it] 30%|███ | 12/40 [01:46<04:04, 8.75s/it] 32%|███▎ | 13/40 [01:55<03:56, 8.75s/it] 35%|███▌ | 14/40 [02:03<03:47, 8.75s/it] 38%|███▊ | 15/40 [02:12<03:38, 8.75s/it] 40%|████ | 16/40 [02:21<03:29, 8.75s/it] 42%|████▎ | 17/40 [02:29<03:21, 8.75s/it] 45%|████▌ | 18/40 [02:38<03:12, 8.75s/it] 48%|████▊ | 19/40 [02:47<03:03, 8.75s/it] 50%|█████ | 20/40 [02:56<02:54, 8.75s/it] 52%|█████▎ | 21/40 [03:04<02:46, 8.75s/it] 55%|█████▌ | 22/40 [03:13<02:37, 8.75s/it] 57%|█████▊ | 23/40 [03:22<02:28, 8.75s/it] 60%|██████ | 24/40 [03:31<02:19, 8.75s/it] 62%|██████▎ | 25/40 [03:39<02:11, 8.75s/it] 65%|██████▌ | 26/40 [03:48<02:02, 8.75s/it] 68%|██████▊ | 27/40 [03:57<01:53, 8.75s/it] 70%|███████ | 28/40 [04:06<01:44, 8.75s/it] 72%|███████▎ | 29/40 [04:14<01:36, 8.76s/it] 75%|███████▌ | 30/40 [04:23<01:27, 8.76s/it] 78%|███████▊ | 31/40 [04:32<01:18, 8.76s/it] 80%|████████ | 32/40 [04:41<01:10, 8.76s/it] 82%|████████▎ | 33/40 [04:50<01:01, 8.76s/it] 85%|████████▌ | 34/40 [04:58<00:52, 8.75s/it] 88%|████████▊ | 35/40 [05:07<00:43, 8.75s/it] 90%|█████████ | 36/40 [05:16<00:35, 8.77s/it] 92%|█████████▎| 37/40 [05:25<00:26, 8.76s/it] 95%|█████████▌| 38/40 [05:33<00:17, 8.76s/it] 98%|█████████▊| 39/40 [05:42<00:08, 8.75s/it] 100%|██████████| 40/40 [05:51<00:00, 8.76s/it] 100%|██████████| 40/40 [05:51<00:00, 8.78s/it] 2024-12-09 11:02:52,817-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.985 GB 2024-12-09 11:02:52,817-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=17.704 GB 2024-12-09 11:02:52,817-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=19.467 GB
Prediction
zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582bID3jwp6t25f1rma0ckncdsbe4mhmStatusSucceededSourceWebHardwareL40STotal durationCreatedInput
- seed
- 43
- width
- 960
- height
- 544
- prompt
- A vibrant, high-quality 4K cinematic scene of a quick brown fox leaping gracefully over a lazy dog lying on a lush, sunlit meadow. The fox's fur is sleek and shines in the sunlight, its motion captured mid-air with detailed muscle definition and dynamic movement. The lazy dog, a large, gentle breed with soft fur, lies peacefully on the grass, its expression calm and undisturbed. The scene is set in a picturesque natural environment with a backdrop of rolling green hills, scattered wildflowers, and a bright blue sky with fluffy white clouds. The moment is captured with vivid detail, showcasing the textures of fur, the gentle sway of grass in the breeze, and the serene beauty of the outdoor setting.
- flow_shift
- 7
- video_length
- 65
- negative_prompt
- Ugly
- num_inference_steps
- 40
- embedded_guidance_scale
- 6
{ "seed": 43, "width": 960, "height": 544, "prompt": "A vibrant, high-quality 4K cinematic scene of a quick brown fox leaping gracefully over a lazy dog lying on a lush, sunlit meadow. The fox's fur is sleek and shines in the sunlight, its motion captured mid-air with detailed muscle definition and dynamic movement. The lazy dog, a large, gentle breed with soft fur, lies peacefully on the grass, its expression calm and undisturbed. The scene is set in a picturesque natural environment with a backdrop of rolling green hills, scattered wildflowers, and a bright blue sky with fluffy white clouds. The moment is captured with vivid detail, showcasing the textures of fur, the gentle sway of grass in the breeze, and the serene beauty of the outdoor setting.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }
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 zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", { input: { seed: 43, width: 960, height: 544, prompt: "A vibrant, high-quality 4K cinematic scene of a quick brown fox leaping gracefully over a lazy dog lying on a lush, sunlit meadow. The fox's fur is sleek and shines in the sunlight, its motion captured mid-air with detailed muscle definition and dynamic movement. The lazy dog, a large, gentle breed with soft fur, lies peacefully on the grass, its expression calm and undisturbed. The scene is set in a picturesque natural environment with a backdrop of rolling green hills, scattered wildflowers, and a bright blue sky with fluffy white clouds. The moment is captured with vivid detail, showcasing the textures of fur, the gentle sway of grass in the breeze, and the serene beauty of the outdoor setting.", flow_shift: 7, video_length: 65, negative_prompt: "Ugly", num_inference_steps: 40, embedded_guidance_scale: 6 } } ); // 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 Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", input={ "seed": 43, "width": 960, "height": 544, "prompt": "A vibrant, high-quality 4K cinematic scene of a quick brown fox leaping gracefully over a lazy dog lying on a lush, sunlit meadow. The fox's fur is sleek and shines in the sunlight, its motion captured mid-air with detailed muscle definition and dynamic movement. The lazy dog, a large, gentle breed with soft fur, lies peacefully on the grass, its expression calm and undisturbed. The scene is set in a picturesque natural environment with a backdrop of rolling green hills, scattered wildflowers, and a bright blue sky with fluffy white clouds. The moment is captured with vivid detail, showcasing the textures of fur, the gentle sway of grass in the breeze, and the serene beauty of the outdoor setting.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } ) print(output)
To learn more, take a look at the guide on getting started with Python.
Run zurk/hunyuan-video-8bit 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": "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A vibrant, high-quality 4K cinematic scene of a quick brown fox leaping gracefully over a lazy dog lying on a lush, sunlit meadow. The fox\'s fur is sleek and shines in the sunlight, its motion captured mid-air with detailed muscle definition and dynamic movement. The lazy dog, a large, gentle breed with soft fur, lies peacefully on the grass, its expression calm and undisturbed. The scene is set in a picturesque natural environment with a backdrop of rolling green hills, scattered wildflowers, and a bright blue sky with fluffy white clouds. The moment is captured with vivid detail, showcasing the textures of fur, the gentle sway of grass in the breeze, and the serene beauty of the outdoor setting.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
Output
{ "completed_at": "2024-12-09T11:15:15.737217Z", "created_at": "2024-12-09T11:08:31.480000Z", "data_removed": false, "error": null, "id": "3jwp6t25f1rma0ckncdsbe4mhm", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A vibrant, high-quality 4K cinematic scene of a quick brown fox leaping gracefully over a lazy dog lying on a lush, sunlit meadow. The fox's fur is sleek and shines in the sunlight, its motion captured mid-air with detailed muscle definition and dynamic movement. The lazy dog, a large, gentle breed with soft fur, lies peacefully on the grass, its expression calm and undisturbed. The scene is set in a picturesque natural environment with a backdrop of rolling green hills, scattered wildflowers, and a bright blue sky with fluffy white clouds. The moment is captured with vivid detail, showcasing the textures of fur, the gentle sway of grass in the breeze, and the serene beauty of the outdoor setting.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }, "logs": "2024-12-09 11:08:50,538-INFO-cog_predictor-seed=43\n2024-12-09 11:08:57,826-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 65)\n2024-12-09 11:09:00,330-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 65 frames in 17 latents at 960x544 with 40 inference steps\n 0%| | 0/40 [00:00<?, ?it/s]\n 2%|▎ | 1/40 [00:10<06:51, 10.56s/it]\n 5%|▌ | 2/40 [00:19<05:57, 9.41s/it]\n 8%|▊ | 3/40 [00:27<05:36, 9.11s/it]\n 10%|█ | 4/40 [00:36<05:22, 8.96s/it]\n 12%|█▎ | 5/40 [00:45<05:11, 8.89s/it]\n 15%|█▌ | 6/40 [00:54<05:00, 8.85s/it]\n 18%|█▊ | 7/40 [01:02<04:51, 8.83s/it]\n 20%|██ | 8/40 [01:11<04:42, 8.82s/it]\n 22%|██▎ | 9/40 [01:20<04:32, 8.80s/it]\n 25%|██▌ | 10/40 [01:29<04:23, 8.79s/it]\n 28%|██▊ | 11/40 [01:38<04:15, 8.79s/it]\n 30%|███ | 12/40 [01:46<04:06, 8.80s/it]\n 32%|███▎ | 13/40 [01:55<03:57, 8.79s/it]\n 35%|███▌ | 14/40 [02:04<03:48, 8.80s/it]\n 38%|███▊ | 15/40 [02:13<03:40, 8.80s/it]\n 40%|████ | 16/40 [02:22<03:31, 8.81s/it]\n 42%|████▎ | 17/40 [02:30<03:22, 8.81s/it]\n 45%|████▌ | 18/40 [02:39<03:13, 8.81s/it]\n 48%|████▊ | 19/40 [02:48<03:05, 8.81s/it]\n 50%|█████ | 20/40 [02:57<02:56, 8.81s/it]\n 52%|█████▎ | 21/40 [03:06<02:47, 8.82s/it]\n 55%|█████▌ | 22/40 [03:15<02:38, 8.82s/it]\n 57%|█████▊ | 23/40 [03:23<02:29, 8.82s/it]\n 60%|██████ | 24/40 [03:32<02:20, 8.81s/it]\n 62%|██████▎ | 25/40 [03:41<02:12, 8.81s/it]\n 65%|██████▌ | 26/40 [03:50<02:03, 8.81s/it]\n 68%|██████▊ | 27/40 [03:59<01:54, 8.80s/it]\n 70%|███████ | 28/40 [04:07<01:45, 8.81s/it]\n 72%|███████▎ | 29/40 [04:16<01:36, 8.81s/it]\n 75%|███████▌ | 30/40 [04:25<01:28, 8.81s/it]\n 78%|███████▊ | 31/40 [04:34<01:19, 8.81s/it]\n 80%|████████ | 32/40 [04:43<01:10, 8.81s/it]\n 82%|████████▎ | 33/40 [04:51<01:01, 8.80s/it]\n 85%|████████▌ | 34/40 [05:00<00:52, 8.80s/it]\n 88%|████████▊ | 35/40 [05:09<00:43, 8.79s/it]\n 90%|█████████ | 36/40 [05:18<00:35, 8.79s/it]\n 92%|█████████▎| 37/40 [05:27<00:26, 8.80s/it]\n 95%|█████████▌| 38/40 [05:35<00:17, 8.79s/it]\n 98%|█████████▊| 39/40 [05:44<00:08, 8.80s/it]\n100%|██████████| 40/40 [05:53<00:00, 8.80s/it]\n100%|██████████| 40/40 [05:53<00:00, 8.84s/it]\n2024-12-09 11:14:53,795-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.985 GB\n2024-12-09 11:14:53,795-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=17.704 GB\n2024-12-09 11:14:53,795-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=19.467 GB", "metrics": { "predict_time": 385.197043667, "total_time": 404.257217 }, "output": "https://replicate.delivery/xezq/HKWhM8YX4l6rMp790N7PsB3ebUdYpaLeyca8XS4AiBeHaQynA/HunyuanVideo_00008.mp4", "started_at": "2024-12-09T11:08:50.540173Z", "status": "succeeded", "urls": { "stream": "https://stream.replicate.com/v1/files/bcwr-jgx7zr7exj2q6ui5q3mxzsony3wrm2eib3cvxb3auts74lf5t4lq", "get": "https://api.replicate.com/v1/predictions/3jwp6t25f1rma0ckncdsbe4mhm", "cancel": "https://api.replicate.com/v1/predictions/3jwp6t25f1rma0ckncdsbe4mhm/cancel" }, "version": "392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b" }
Generated in2024-12-09 11:08:50,538-INFO-cog_predictor-seed=43 2024-12-09 11:08:57,826-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 65) 2024-12-09 11:09:00,330-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 65 frames in 17 latents at 960x544 with 40 inference steps 0%| | 0/40 [00:00<?, ?it/s] 2%|▎ | 1/40 [00:10<06:51, 10.56s/it] 5%|▌ | 2/40 [00:19<05:57, 9.41s/it] 8%|▊ | 3/40 [00:27<05:36, 9.11s/it] 10%|█ | 4/40 [00:36<05:22, 8.96s/it] 12%|█▎ | 5/40 [00:45<05:11, 8.89s/it] 15%|█▌ | 6/40 [00:54<05:00, 8.85s/it] 18%|█▊ | 7/40 [01:02<04:51, 8.83s/it] 20%|██ | 8/40 [01:11<04:42, 8.82s/it] 22%|██▎ | 9/40 [01:20<04:32, 8.80s/it] 25%|██▌ | 10/40 [01:29<04:23, 8.79s/it] 28%|██▊ | 11/40 [01:38<04:15, 8.79s/it] 30%|███ | 12/40 [01:46<04:06, 8.80s/it] 32%|███▎ | 13/40 [01:55<03:57, 8.79s/it] 35%|███▌ | 14/40 [02:04<03:48, 8.80s/it] 38%|███▊ | 15/40 [02:13<03:40, 8.80s/it] 40%|████ | 16/40 [02:22<03:31, 8.81s/it] 42%|████▎ | 17/40 [02:30<03:22, 8.81s/it] 45%|████▌ | 18/40 [02:39<03:13, 8.81s/it] 48%|████▊ | 19/40 [02:48<03:05, 8.81s/it] 50%|█████ | 20/40 [02:57<02:56, 8.81s/it] 52%|█████▎ | 21/40 [03:06<02:47, 8.82s/it] 55%|█████▌ | 22/40 [03:15<02:38, 8.82s/it] 57%|█████▊ | 23/40 [03:23<02:29, 8.82s/it] 60%|██████ | 24/40 [03:32<02:20, 8.81s/it] 62%|██████▎ | 25/40 [03:41<02:12, 8.81s/it] 65%|██████▌ | 26/40 [03:50<02:03, 8.81s/it] 68%|██████▊ | 27/40 [03:59<01:54, 8.80s/it] 70%|███████ | 28/40 [04:07<01:45, 8.81s/it] 72%|███████▎ | 29/40 [04:16<01:36, 8.81s/it] 75%|███████▌ | 30/40 [04:25<01:28, 8.81s/it] 78%|███████▊ | 31/40 [04:34<01:19, 8.81s/it] 80%|████████ | 32/40 [04:43<01:10, 8.81s/it] 82%|████████▎ | 33/40 [04:51<01:01, 8.80s/it] 85%|████████▌ | 34/40 [05:00<00:52, 8.80s/it] 88%|████████▊ | 35/40 [05:09<00:43, 8.79s/it] 90%|█████████ | 36/40 [05:18<00:35, 8.79s/it] 92%|█████████▎| 37/40 [05:27<00:26, 8.80s/it] 95%|█████████▌| 38/40 [05:35<00:17, 8.79s/it] 98%|█████████▊| 39/40 [05:44<00:08, 8.80s/it] 100%|██████████| 40/40 [05:53<00:00, 8.80s/it] 100%|██████████| 40/40 [05:53<00:00, 8.84s/it] 2024-12-09 11:14:53,795-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.985 GB 2024-12-09 11:14:53,795-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=17.704 GB 2024-12-09 11:14:53,795-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=19.467 GB
Prediction
zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582bID6a1sgmv5bxrme0ckncj91gkr5cStatusSucceededSourceWebHardwareL40STotal durationCreatedInput
- seed
- 43
- width
- 960
- height
- 544
- prompt
- A cinematic scene of a person walking on a rainy night in a bustling New York City street. The person is dressed in a long coat and holding an umbrella, their silhouette illuminated by the colorful reflections of neon signs and streetlights shimmering on the wet pavement. The rain softly falls, creating ripples in small puddles on the ground. The atmosphere is vibrant yet moody, with passing cars, blurred lights, and faint steam rising from street grates. The background captures iconic New York elements, such as fire escapes, busy sidewalks, and yellow cabs. The video is in 4K resolution, emphasizing the fine details of raindrops, the texture of the person’s coat, and the vivid city lights.
- flow_shift
- 7
- video_length
- 65
- negative_prompt
- Ugly
- num_inference_steps
- 40
- embedded_guidance_scale
- 6
{ "seed": 43, "width": 960, "height": 544, "prompt": "A cinematic scene of a person walking on a rainy night in a bustling New York City street. The person is dressed in a long coat and holding an umbrella, their silhouette illuminated by the colorful reflections of neon signs and streetlights shimmering on the wet pavement. The rain softly falls, creating ripples in small puddles on the ground. The atmosphere is vibrant yet moody, with passing cars, blurred lights, and faint steam rising from street grates. The background captures iconic New York elements, such as fire escapes, busy sidewalks, and yellow cabs. The video is in 4K resolution, emphasizing the fine details of raindrops, the texture of the person’s coat, and the vivid city lights.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }
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 zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", { input: { seed: 43, width: 960, height: 544, prompt: "A cinematic scene of a person walking on a rainy night in a bustling New York City street. The person is dressed in a long coat and holding an umbrella, their silhouette illuminated by the colorful reflections of neon signs and streetlights shimmering on the wet pavement. The rain softly falls, creating ripples in small puddles on the ground. The atmosphere is vibrant yet moody, with passing cars, blurred lights, and faint steam rising from street grates. The background captures iconic New York elements, such as fire escapes, busy sidewalks, and yellow cabs. The video is in 4K resolution, emphasizing the fine details of raindrops, the texture of the person’s coat, and the vivid city lights.", flow_shift: 7, video_length: 65, negative_prompt: "Ugly", num_inference_steps: 40, embedded_guidance_scale: 6 } } ); // 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 Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", input={ "seed": 43, "width": 960, "height": 544, "prompt": "A cinematic scene of a person walking on a rainy night in a bustling New York City street. The person is dressed in a long coat and holding an umbrella, their silhouette illuminated by the colorful reflections of neon signs and streetlights shimmering on the wet pavement. The rain softly falls, creating ripples in small puddles on the ground. The atmosphere is vibrant yet moody, with passing cars, blurred lights, and faint steam rising from street grates. The background captures iconic New York elements, such as fire escapes, busy sidewalks, and yellow cabs. The video is in 4K resolution, emphasizing the fine details of raindrops, the texture of the person’s coat, and the vivid city lights.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } ) print(output)
To learn more, take a look at the guide on getting started with Python.
Run zurk/hunyuan-video-8bit 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": "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A cinematic scene of a person walking on a rainy night in a bustling New York City street. The person is dressed in a long coat and holding an umbrella, their silhouette illuminated by the colorful reflections of neon signs and streetlights shimmering on the wet pavement. The rain softly falls, creating ripples in small puddles on the ground. The atmosphere is vibrant yet moody, with passing cars, blurred lights, and faint steam rising from street grates. The background captures iconic New York elements, such as fire escapes, busy sidewalks, and yellow cabs. The video is in 4K resolution, emphasizing the fine details of raindrops, the texture of the person’s coat, and the vivid city lights.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
Output
{ "completed_at": "2024-12-09T11:28:51.787247Z", "created_at": "2024-12-09T11:18:29.471000Z", "data_removed": false, "error": null, "id": "6a1sgmv5bxrme0ckncj91gkr5c", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A cinematic scene of a person walking on a rainy night in a bustling New York City street. The person is dressed in a long coat and holding an umbrella, their silhouette illuminated by the colorful reflections of neon signs and streetlights shimmering on the wet pavement. The rain softly falls, creating ripples in small puddles on the ground. The atmosphere is vibrant yet moody, with passing cars, blurred lights, and faint steam rising from street grates. The background captures iconic New York elements, such as fire escapes, busy sidewalks, and yellow cabs. The video is in 4K resolution, emphasizing the fine details of raindrops, the texture of the person’s coat, and the vivid city lights.", "flow_shift": 7, "video_length": 65, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }, "logs": "2024-12-09 11:22:21,449-INFO-cog_predictor-seed=43\n2024-12-09 11:22:28,902-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 65)\n2024-12-09 11:22:31,211-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 65 frames in 17 latents at 960x544 with 40 inference steps\n 0%| | 0/40 [00:00<?, ?it/s]\n 2%|▎ | 1/40 [00:10<06:54, 10.63s/it]\n 5%|▌ | 2/40 [00:19<06:00, 9.48s/it]\n 8%|▊ | 3/40 [00:28<05:38, 9.16s/it]\n 10%|█ | 4/40 [00:36<05:24, 9.02s/it]\n 12%|█▎ | 5/40 [00:45<05:12, 8.94s/it]\n 15%|█▌ | 6/40 [00:54<05:02, 8.90s/it]\n 18%|█▊ | 7/40 [01:03<04:53, 8.90s/it]\n 20%|██ | 8/40 [01:12<04:44, 8.88s/it]\n 22%|██▎ | 9/40 [01:21<04:35, 8.88s/it]\n 25%|██▌ | 10/40 [01:30<04:26, 8.88s/it]\n 28%|██▊ | 11/40 [01:38<04:17, 8.87s/it]\n 30%|███ | 12/40 [01:47<04:08, 8.89s/it]\n 32%|███▎ | 13/40 [01:56<04:00, 8.89s/it]\n 35%|███▌ | 14/40 [02:05<03:51, 8.91s/it]\n 38%|███▊ | 15/40 [02:14<03:43, 8.92s/it]\n 40%|████ | 16/40 [02:23<03:34, 8.94s/it]\n 42%|████▎ | 17/40 [02:32<03:25, 8.95s/it]\n 45%|████▌ | 18/40 [02:41<03:17, 8.97s/it]\n 48%|████▊ | 19/40 [02:50<03:08, 8.97s/it]\n 50%|█████ | 20/40 [02:59<02:59, 8.98s/it]\n 52%|█████▎ | 21/40 [03:08<02:50, 8.98s/it]\n 55%|█████▌ | 22/40 [03:17<02:41, 8.99s/it]\n 57%|█████▊ | 23/40 [03:26<02:32, 8.98s/it]\n 60%|██████ | 24/40 [03:35<02:23, 8.98s/it]\n 62%|██████▎ | 25/40 [03:44<02:14, 8.99s/it]\n 65%|██████▌ | 26/40 [03:53<02:05, 8.98s/it]\n 68%|██████▊ | 27/40 [04:02<01:56, 8.98s/it]\n 70%|███████ | 28/40 [04:11<01:47, 8.96s/it]\n 72%|███████▎ | 29/40 [04:20<01:38, 8.96s/it]\n 75%|███████▌ | 30/40 [04:29<01:29, 8.95s/it]\n 78%|███████▊ | 31/40 [04:38<01:20, 8.95s/it]\n 80%|████████ | 32/40 [04:47<01:11, 8.96s/it]\n 82%|████████▎ | 33/40 [04:56<01:02, 8.96s/it]\n 85%|████████▌ | 34/40 [05:05<00:53, 8.96s/it]\n 88%|████████▊ | 35/40 [05:14<00:44, 8.96s/it]\n 90%|█████████ | 36/40 [05:22<00:35, 8.96s/it]\n 92%|█████████▎| 37/40 [05:31<00:26, 8.95s/it]\n 95%|█████████▌| 38/40 [05:40<00:17, 8.95s/it]\n 98%|█████████▊| 39/40 [05:49<00:08, 8.96s/it]\n100%|██████████| 40/40 [05:58<00:00, 8.96s/it]\n100%|██████████| 40/40 [05:58<00:00, 8.97s/it]\n2024-12-09 11:28:29,992-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.985 GB\n2024-12-09 11:28:29,992-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=17.704 GB\n2024-12-09 11:28:29,992-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=19.467 GB", "metrics": { "predict_time": 390.335301154, "total_time": 622.316247 }, "output": "https://replicate.delivery/xezq/TewrraGmDhzgA6vfI4BZ8MzyLuqZdvOfdUd5tnCmM10nzQynA/HunyuanVideo_00008.mp4", "started_at": "2024-12-09T11:22:21.451946Z", "status": "succeeded", "urls": { "stream": "https://stream.replicate.com/v1/files/bcwr-r3aexwttdczwmyvp7szlddnam3uhpdwnaqhmospdyw7djoy5nl6a", "get": "https://api.replicate.com/v1/predictions/6a1sgmv5bxrme0ckncj91gkr5c", "cancel": "https://api.replicate.com/v1/predictions/6a1sgmv5bxrme0ckncj91gkr5c/cancel" }, "version": "392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b" }
Generated in2024-12-09 11:22:21,449-INFO-cog_predictor-seed=43 2024-12-09 11:22:28,902-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 65) 2024-12-09 11:22:31,211-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 65 frames in 17 latents at 960x544 with 40 inference steps 0%| | 0/40 [00:00<?, ?it/s] 2%|▎ | 1/40 [00:10<06:54, 10.63s/it] 5%|▌ | 2/40 [00:19<06:00, 9.48s/it] 8%|▊ | 3/40 [00:28<05:38, 9.16s/it] 10%|█ | 4/40 [00:36<05:24, 9.02s/it] 12%|█▎ | 5/40 [00:45<05:12, 8.94s/it] 15%|█▌ | 6/40 [00:54<05:02, 8.90s/it] 18%|█▊ | 7/40 [01:03<04:53, 8.90s/it] 20%|██ | 8/40 [01:12<04:44, 8.88s/it] 22%|██▎ | 9/40 [01:21<04:35, 8.88s/it] 25%|██▌ | 10/40 [01:30<04:26, 8.88s/it] 28%|██▊ | 11/40 [01:38<04:17, 8.87s/it] 30%|███ | 12/40 [01:47<04:08, 8.89s/it] 32%|███▎ | 13/40 [01:56<04:00, 8.89s/it] 35%|███▌ | 14/40 [02:05<03:51, 8.91s/it] 38%|███▊ | 15/40 [02:14<03:43, 8.92s/it] 40%|████ | 16/40 [02:23<03:34, 8.94s/it] 42%|████▎ | 17/40 [02:32<03:25, 8.95s/it] 45%|████▌ | 18/40 [02:41<03:17, 8.97s/it] 48%|████▊ | 19/40 [02:50<03:08, 8.97s/it] 50%|█████ | 20/40 [02:59<02:59, 8.98s/it] 52%|█████▎ | 21/40 [03:08<02:50, 8.98s/it] 55%|█████▌ | 22/40 [03:17<02:41, 8.99s/it] 57%|█████▊ | 23/40 [03:26<02:32, 8.98s/it] 60%|██████ | 24/40 [03:35<02:23, 8.98s/it] 62%|██████▎ | 25/40 [03:44<02:14, 8.99s/it] 65%|██████▌ | 26/40 [03:53<02:05, 8.98s/it] 68%|██████▊ | 27/40 [04:02<01:56, 8.98s/it] 70%|███████ | 28/40 [04:11<01:47, 8.96s/it] 72%|███████▎ | 29/40 [04:20<01:38, 8.96s/it] 75%|███████▌ | 30/40 [04:29<01:29, 8.95s/it] 78%|███████▊ | 31/40 [04:38<01:20, 8.95s/it] 80%|████████ | 32/40 [04:47<01:11, 8.96s/it] 82%|████████▎ | 33/40 [04:56<01:02, 8.96s/it] 85%|████████▌ | 34/40 [05:05<00:53, 8.96s/it] 88%|████████▊ | 35/40 [05:14<00:44, 8.96s/it] 90%|█████████ | 36/40 [05:22<00:35, 8.96s/it] 92%|█████████▎| 37/40 [05:31<00:26, 8.95s/it] 95%|█████████▌| 38/40 [05:40<00:17, 8.95s/it] 98%|█████████▊| 39/40 [05:49<00:08, 8.96s/it] 100%|██████████| 40/40 [05:58<00:00, 8.96s/it] 100%|██████████| 40/40 [05:58<00:00, 8.97s/it] 2024-12-09 11:28:29,992-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.985 GB 2024-12-09 11:28:29,992-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=17.704 GB 2024-12-09 11:28:29,992-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=19.467 GB
Prediction
zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582bID3ycmb5nxbnrmc0cknd1vgza9r8StatusSucceededSourceWebHardwareL40STotal durationCreatedInput
- seed
- 43
- width
- 960
- height
- 544
- prompt
- A high-quality, cinematic anime-style depiction of a serene scene featuring a young anime girl. She stands on a grassy hill, her long hair flowing gracefully in the gentle breeze. The art style is inspired by a blend of modern anime aesthetics with vibrant colors and soft, painterly shading. She wears a flowing summer dress, detailed with intricate patterns, that moves lightly with the wind. The background is a picturesque landscape with rolling hills, blooming wildflowers, and a golden sunset casting a warm glow over the scene. Her large, expressive eyes capture a sense of wonder and calm, embodying the essence of classic and modern anime artistry.
- flow_shift
- 7
- video_length
- 97
- negative_prompt
- Ugly
- num_inference_steps
- 40
- embedded_guidance_scale
- 6
{ "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic anime-style depiction of a serene scene featuring a young anime girl. She stands on a grassy hill, her long hair flowing gracefully in the gentle breeze. The art style is inspired by a blend of modern anime aesthetics with vibrant colors and soft, painterly shading. She wears a flowing summer dress, detailed with intricate patterns, that moves lightly with the wind. The background is a picturesque landscape with rolling hills, blooming wildflowers, and a golden sunset casting a warm glow over the scene. Her large, expressive eyes capture a sense of wonder and calm, embodying the essence of classic and modern anime artistry.", "flow_shift": 7, "video_length": 97, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }
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 zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", { input: { seed: 43, width: 960, height: 544, prompt: "A high-quality, cinematic anime-style depiction of a serene scene featuring a young anime girl. She stands on a grassy hill, her long hair flowing gracefully in the gentle breeze. The art style is inspired by a blend of modern anime aesthetics with vibrant colors and soft, painterly shading. She wears a flowing summer dress, detailed with intricate patterns, that moves lightly with the wind. The background is a picturesque landscape with rolling hills, blooming wildflowers, and a golden sunset casting a warm glow over the scene. Her large, expressive eyes capture a sense of wonder and calm, embodying the essence of classic and modern anime artistry.", flow_shift: 7, video_length: 97, negative_prompt: "Ugly", num_inference_steps: 40, embedded_guidance_scale: 6 } } ); // 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 Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", input={ "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic anime-style depiction of a serene scene featuring a young anime girl. She stands on a grassy hill, her long hair flowing gracefully in the gentle breeze. The art style is inspired by a blend of modern anime aesthetics with vibrant colors and soft, painterly shading. She wears a flowing summer dress, detailed with intricate patterns, that moves lightly with the wind. The background is a picturesque landscape with rolling hills, blooming wildflowers, and a golden sunset casting a warm glow over the scene. Her large, expressive eyes capture a sense of wonder and calm, embodying the essence of classic and modern anime artistry.", "flow_shift": 7, "video_length": 97, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } ) print(output)
To learn more, take a look at the guide on getting started with Python.
Run zurk/hunyuan-video-8bit 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": "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic anime-style depiction of a serene scene featuring a young anime girl. She stands on a grassy hill, her long hair flowing gracefully in the gentle breeze. The art style is inspired by a blend of modern anime aesthetics with vibrant colors and soft, painterly shading. She wears a flowing summer dress, detailed with intricate patterns, that moves lightly with the wind. The background is a picturesque landscape with rolling hills, blooming wildflowers, and a golden sunset casting a warm glow over the scene. Her large, expressive eyes capture a sense of wonder and calm, embodying the essence of classic and modern anime artistry.", "flow_shift": 7, "video_length": 97, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
Output
{ "completed_at": "2024-12-09T12:08:13.640542Z", "created_at": "2024-12-09T11:52:43.613000Z", "data_removed": false, "error": null, "id": "3ycmb5nxbnrmc0cknd1vgza9r8", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic anime-style depiction of a serene scene featuring a young anime girl. She stands on a grassy hill, her long hair flowing gracefully in the gentle breeze. The art style is inspired by a blend of modern anime aesthetics with vibrant colors and soft, painterly shading. She wears a flowing summer dress, detailed with intricate patterns, that moves lightly with the wind. The background is a picturesque landscape with rolling hills, blooming wildflowers, and a golden sunset casting a warm glow over the scene. Her large, expressive eyes capture a sense of wonder and calm, embodying the essence of classic and modern anime artistry.", "flow_shift": 7, "video_length": 97, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }, "logs": "2024-12-09 11:57:37,970-INFO-cog_predictor-seed=43\n2024-12-09 11:57:45,239-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 97)\n2024-12-09 11:57:47,632-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 97 frames in 25 latents at 960x544 with 40 inference steps\n 0%| | 0/40 [00:00<?, ?it/s]\n 2%|▎ | 1/40 [00:16<10:44, 16.52s/it]\n 5%|▌ | 2/40 [00:31<09:45, 15.40s/it]\n 8%|▊ | 3/40 [00:45<09:20, 15.14s/it]\n 10%|█ | 4/40 [01:00<09:01, 15.05s/it]\n 12%|█▎ | 5/40 [01:15<08:44, 14.99s/it]\n 15%|█▌ | 6/40 [01:30<08:28, 14.97s/it]\n 18%|█▊ | 7/40 [01:45<08:13, 14.96s/it]\n 20%|██ | 8/40 [02:00<07:58, 14.95s/it]\n 22%|██▎ | 9/40 [02:15<07:43, 14.95s/it]\n 25%|██▌ | 10/40 [02:30<07:28, 14.95s/it]\n 28%|██▊ | 11/40 [02:45<07:13, 14.95s/it]\n 30%|███ | 12/40 [03:00<06:58, 14.96s/it]\n 32%|███▎ | 13/40 [03:15<06:43, 14.95s/it]\n 35%|███▌ | 14/40 [03:30<06:28, 14.94s/it]\n 38%|███▊ | 15/40 [03:45<06:13, 14.95s/it]\n 40%|████ | 16/40 [04:00<05:58, 14.96s/it]\n 42%|████▎ | 17/40 [04:15<05:44, 14.96s/it]\n 45%|████▌ | 18/40 [04:30<05:29, 14.96s/it]\n 48%|████▊ | 19/40 [04:45<05:14, 14.97s/it]\n 50%|█████ | 20/40 [05:00<04:59, 14.97s/it]\n 52%|█████▎ | 21/40 [05:15<04:44, 14.96s/it]\n 55%|█████▌ | 22/40 [05:29<04:29, 14.96s/it]\n 57%|█████▊ | 23/40 [05:44<04:14, 14.94s/it]\n 60%|██████ | 24/40 [05:59<03:58, 14.93s/it]\n 62%|██████▎ | 25/40 [06:14<03:43, 14.93s/it]\n 65%|██████▌ | 26/40 [06:29<03:29, 14.94s/it]\n 68%|██████▊ | 27/40 [06:44<03:14, 14.93s/it]\n 70%|███████ | 28/40 [06:59<02:59, 14.92s/it]\n 72%|███████▎ | 29/40 [07:14<02:44, 14.93s/it]\n 75%|███████▌ | 30/40 [07:29<02:29, 14.94s/it]\n 78%|███████▊ | 31/40 [07:44<02:14, 14.93s/it]\n 80%|████████ | 32/40 [07:59<01:59, 14.93s/it]\n 82%|████████▎ | 33/40 [08:14<01:44, 14.94s/it]\n 85%|████████▌ | 34/40 [08:29<01:29, 14.93s/it]\n 88%|████████▊ | 35/40 [08:44<01:14, 14.93s/it]\n 90%|█████████ | 36/40 [08:58<00:59, 14.93s/it]\n 92%|█████████▎| 37/40 [09:13<00:44, 14.92s/it]\n 95%|█████████▌| 38/40 [09:28<00:29, 14.92s/it]\n 98%|█████████▊| 39/40 [09:43<00:14, 14.92s/it]\n100%|██████████| 40/40 [09:58<00:00, 14.92s/it]\n100%|██████████| 40/40 [09:58<00:00, 14.97s/it]\n2024-12-09 12:07:46,241-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.989 GB\n2024-12-09 12:07:46,241-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=20.325 GB\n2024-12-09 12:07:46,241-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=23.021 GB", "metrics": { "predict_time": 635.667367889, "total_time": 930.027542 }, "output": "https://replicate.delivery/xezq/fdNGeOoe1ZKv8IyLgEOokPbvWNWWkWejQrVV0fXdv6nu1HJfE/HunyuanVideo_00008.mp4", "started_at": "2024-12-09T11:57:37.973174Z", "status": "succeeded", "urls": { "stream": "https://stream.replicate.com/v1/files/bcwr-dshebkjf42tfnkkp5wwk4h4qib3n3mqksfqvwjggfm5j6oklcw2a", "get": "https://api.replicate.com/v1/predictions/3ycmb5nxbnrmc0cknd1vgza9r8", "cancel": "https://api.replicate.com/v1/predictions/3ycmb5nxbnrmc0cknd1vgza9r8/cancel" }, "version": "392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b" }
Generated in2024-12-09 11:57:37,970-INFO-cog_predictor-seed=43 2024-12-09 11:57:45,239-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 97) 2024-12-09 11:57:47,632-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 97 frames in 25 latents at 960x544 with 40 inference steps 0%| | 0/40 [00:00<?, ?it/s] 2%|▎ | 1/40 [00:16<10:44, 16.52s/it] 5%|▌ | 2/40 [00:31<09:45, 15.40s/it] 8%|▊ | 3/40 [00:45<09:20, 15.14s/it] 10%|█ | 4/40 [01:00<09:01, 15.05s/it] 12%|█▎ | 5/40 [01:15<08:44, 14.99s/it] 15%|█▌ | 6/40 [01:30<08:28, 14.97s/it] 18%|█▊ | 7/40 [01:45<08:13, 14.96s/it] 20%|██ | 8/40 [02:00<07:58, 14.95s/it] 22%|██▎ | 9/40 [02:15<07:43, 14.95s/it] 25%|██▌ | 10/40 [02:30<07:28, 14.95s/it] 28%|██▊ | 11/40 [02:45<07:13, 14.95s/it] 30%|███ | 12/40 [03:00<06:58, 14.96s/it] 32%|███▎ | 13/40 [03:15<06:43, 14.95s/it] 35%|███▌ | 14/40 [03:30<06:28, 14.94s/it] 38%|███▊ | 15/40 [03:45<06:13, 14.95s/it] 40%|████ | 16/40 [04:00<05:58, 14.96s/it] 42%|████▎ | 17/40 [04:15<05:44, 14.96s/it] 45%|████▌ | 18/40 [04:30<05:29, 14.96s/it] 48%|████▊ | 19/40 [04:45<05:14, 14.97s/it] 50%|█████ | 20/40 [05:00<04:59, 14.97s/it] 52%|█████▎ | 21/40 [05:15<04:44, 14.96s/it] 55%|█████▌ | 22/40 [05:29<04:29, 14.96s/it] 57%|█████▊ | 23/40 [05:44<04:14, 14.94s/it] 60%|██████ | 24/40 [05:59<03:58, 14.93s/it] 62%|██████▎ | 25/40 [06:14<03:43, 14.93s/it] 65%|██████▌ | 26/40 [06:29<03:29, 14.94s/it] 68%|██████▊ | 27/40 [06:44<03:14, 14.93s/it] 70%|███████ | 28/40 [06:59<02:59, 14.92s/it] 72%|███████▎ | 29/40 [07:14<02:44, 14.93s/it] 75%|███████▌ | 30/40 [07:29<02:29, 14.94s/it] 78%|███████▊ | 31/40 [07:44<02:14, 14.93s/it] 80%|████████ | 32/40 [07:59<01:59, 14.93s/it] 82%|████████▎ | 33/40 [08:14<01:44, 14.94s/it] 85%|████████▌ | 34/40 [08:29<01:29, 14.93s/it] 88%|████████▊ | 35/40 [08:44<01:14, 14.93s/it] 90%|█████████ | 36/40 [08:58<00:59, 14.93s/it] 92%|█████████▎| 37/40 [09:13<00:44, 14.92s/it] 95%|█████████▌| 38/40 [09:28<00:29, 14.92s/it] 98%|█████████▊| 39/40 [09:43<00:14, 14.92s/it] 100%|██████████| 40/40 [09:58<00:00, 14.92s/it] 100%|██████████| 40/40 [09:58<00:00, 14.97s/it] 2024-12-09 12:07:46,241-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.989 GB 2024-12-09 12:07:46,241-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=20.325 GB 2024-12-09 12:07:46,241-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=23.021 GB
Prediction
zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582bIDdt9cm2jnahrme0ckndkvwzc5zgStatusSucceededSourceWebHardwareL40STotal durationCreatedInput
- seed
- 43
- width
- 960
- height
- 544
- prompt
- A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.
- flow_shift
- 7
- video_length
- 129
- negative_prompt
- Ugly
- num_inference_steps
- 40
- embedded_guidance_scale
- 6
{ "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.", "flow_shift": 7, "video_length": 129, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }
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 zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", { input: { seed: 43, width: 960, height: 544, prompt: "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.", flow_shift: 7, video_length: 129, negative_prompt: "Ugly", num_inference_steps: 40, embedded_guidance_scale: 6 } } ); // 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 Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", input={ "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.", "flow_shift": 7, "video_length": 129, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } ) print(output)
To learn more, take a look at the guide on getting started with Python.
Run zurk/hunyuan-video-8bit 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": "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli\'s whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane\'s sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki\'s enchanting world.", "flow_shift": 7, "video_length": 129, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
Output
{ "completed_at": "2024-12-09T12:52:19.369236Z", "created_at": "2024-12-09T12:31:36.276000Z", "data_removed": false, "error": null, "id": "dt9cm2jnahrme0ckndkvwzc5zg", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.", "flow_shift": 7, "video_length": 129, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }, "logs": "2024-12-09 12:36:19,069-INFO-cog_predictor-seed=43\n2024-12-09 12:36:26,516-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 129)\n2024-12-09 12:36:28,928-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 129 frames in 33 latents at 960x544 with 40 inference steps\n 0%| | 0/40 [00:00<?, ?it/s]\n 2%|▎ | 1/40 [00:24<15:48, 24.32s/it]\n 5%|▌ | 2/40 [00:46<14:40, 23.17s/it]\n 8%|▊ | 3/40 [01:09<14:09, 22.97s/it]\n 10%|█ | 4/40 [01:32<13:44, 22.89s/it]\n 12%|█▎ | 5/40 [01:55<13:20, 22.87s/it]\n 15%|█▌ | 6/40 [02:17<12:57, 22.88s/it]\n 18%|█▊ | 7/40 [02:40<12:35, 22.88s/it]\n 20%|██ | 8/40 [03:03<12:12, 22.89s/it]\n 22%|██▎ | 9/40 [03:26<11:50, 22.90s/it]\n 25%|██▌ | 10/40 [03:49<11:27, 22.90s/it]\n 28%|██▊ | 11/40 [04:12<11:04, 22.90s/it]\n 30%|███ | 12/40 [04:35<10:41, 22.90s/it]\n 32%|███▎ | 13/40 [04:58<10:18, 22.89s/it]\n 35%|███▌ | 14/40 [05:21<09:55, 22.89s/it]\n 38%|███▊ | 15/40 [05:43<09:31, 22.88s/it]\n 40%|████ | 16/40 [06:06<09:08, 22.87s/it]\n 42%|████▎ | 17/40 [06:29<08:45, 22.87s/it]\n 45%|████▌ | 18/40 [06:52<08:22, 22.86s/it]\n 48%|████▊ | 19/40 [07:15<07:59, 22.85s/it]\n 50%|█████ | 20/40 [07:38<07:36, 22.85s/it]\n 52%|█████▎ | 21/40 [08:01<07:14, 22.86s/it]\n 55%|█████▌ | 22/40 [08:23<06:51, 22.86s/it]\n 57%|█████▊ | 23/40 [08:46<06:28, 22.85s/it]\n 60%|██████ | 24/40 [09:09<06:05, 22.86s/it]\n 62%|██████▎ | 25/40 [09:32<05:42, 22.85s/it]\n 65%|██████▌ | 26/40 [09:55<05:19, 22.85s/it]\n 68%|██████▊ | 27/40 [10:18<04:56, 22.84s/it]\n 70%|███████ | 28/40 [10:40<04:33, 22.83s/it]\n 72%|███████▎ | 29/40 [11:03<04:11, 22.83s/it]\n 75%|███████▌ | 30/40 [11:26<03:48, 22.83s/it]\n 78%|███████▊ | 31/40 [11:49<03:25, 22.83s/it]\n 80%|████████ | 32/40 [12:12<03:02, 22.82s/it]\n 82%|████████▎ | 33/40 [12:35<02:39, 22.82s/it]\n 85%|████████▌ | 34/40 [12:57<02:16, 22.80s/it]\n 88%|████████▊ | 35/40 [13:20<01:53, 22.79s/it]\n 90%|█████████ | 36/40 [13:43<01:31, 22.78s/it]\n 92%|█████████▎| 37/40 [14:06<01:08, 22.77s/it]\n 95%|█████████▌| 38/40 [14:28<00:45, 22.78s/it]\n 98%|█████████▊| 39/40 [14:51<00:22, 22.77s/it]\n100%|██████████| 40/40 [15:14<00:00, 22.76s/it]\n100%|██████████| 40/40 [15:14<00:00, 22.86s/it]\n2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.993 GB\n2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=22.945 GB\n2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=24.859 GB", "metrics": { "predict_time": 960.297328229, "total_time": 1243.093236 }, "output": "https://replicate.delivery/xezq/0ewFiG7Zmt3ZMy42ofOJHAM0zKRSyVNSGcqUqePGJWnGQTynA/HunyuanVideo_00008.mp4", "started_at": "2024-12-09T12:36:19.071908Z", "status": "succeeded", "urls": { "stream": "https://stream.replicate.com/v1/files/bcwr-pbfqlv57lapek5wimofoj6kib5n2jkl77ryjwppj3sbkxk2ai6ta", "get": "https://api.replicate.com/v1/predictions/dt9cm2jnahrme0ckndkvwzc5zg", "cancel": "https://api.replicate.com/v1/predictions/dt9cm2jnahrme0ckndkvwzc5zg/cancel" }, "version": "392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b" }
Generated in2024-12-09 12:36:19,069-INFO-cog_predictor-seed=43 2024-12-09 12:36:26,516-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 129) 2024-12-09 12:36:28,928-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 129 frames in 33 latents at 960x544 with 40 inference steps 0%| | 0/40 [00:00<?, ?it/s] 2%|▎ | 1/40 [00:24<15:48, 24.32s/it] 5%|▌ | 2/40 [00:46<14:40, 23.17s/it] 8%|▊ | 3/40 [01:09<14:09, 22.97s/it] 10%|█ | 4/40 [01:32<13:44, 22.89s/it] 12%|█▎ | 5/40 [01:55<13:20, 22.87s/it] 15%|█▌ | 6/40 [02:17<12:57, 22.88s/it] 18%|█▊ | 7/40 [02:40<12:35, 22.88s/it] 20%|██ | 8/40 [03:03<12:12, 22.89s/it] 22%|██▎ | 9/40 [03:26<11:50, 22.90s/it] 25%|██▌ | 10/40 [03:49<11:27, 22.90s/it] 28%|██▊ | 11/40 [04:12<11:04, 22.90s/it] 30%|███ | 12/40 [04:35<10:41, 22.90s/it] 32%|███▎ | 13/40 [04:58<10:18, 22.89s/it] 35%|███▌ | 14/40 [05:21<09:55, 22.89s/it] 38%|███▊ | 15/40 [05:43<09:31, 22.88s/it] 40%|████ | 16/40 [06:06<09:08, 22.87s/it] 42%|████▎ | 17/40 [06:29<08:45, 22.87s/it] 45%|████▌ | 18/40 [06:52<08:22, 22.86s/it] 48%|████▊ | 19/40 [07:15<07:59, 22.85s/it] 50%|█████ | 20/40 [07:38<07:36, 22.85s/it] 52%|█████▎ | 21/40 [08:01<07:14, 22.86s/it] 55%|█████▌ | 22/40 [08:23<06:51, 22.86s/it] 57%|█████▊ | 23/40 [08:46<06:28, 22.85s/it] 60%|██████ | 24/40 [09:09<06:05, 22.86s/it] 62%|██████▎ | 25/40 [09:32<05:42, 22.85s/it] 65%|██████▌ | 26/40 [09:55<05:19, 22.85s/it] 68%|██████▊ | 27/40 [10:18<04:56, 22.84s/it] 70%|███████ | 28/40 [10:40<04:33, 22.83s/it] 72%|███████▎ | 29/40 [11:03<04:11, 22.83s/it] 75%|███████▌ | 30/40 [11:26<03:48, 22.83s/it] 78%|███████▊ | 31/40 [11:49<03:25, 22.83s/it] 80%|████████ | 32/40 [12:12<03:02, 22.82s/it] 82%|████████▎ | 33/40 [12:35<02:39, 22.82s/it] 85%|████████▌ | 34/40 [12:57<02:16, 22.80s/it] 88%|████████▊ | 35/40 [13:20<01:53, 22.79s/it] 90%|█████████ | 36/40 [13:43<01:31, 22.78s/it] 92%|█████████▎| 37/40 [14:06<01:08, 22.77s/it] 95%|█████████▌| 38/40 [14:28<00:45, 22.78s/it] 98%|█████████▊| 39/40 [14:51<00:22, 22.77s/it] 100%|██████████| 40/40 [15:14<00:00, 22.76s/it] 100%|██████████| 40/40 [15:14<00:00, 22.86s/it] 2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.993 GB 2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=22.945 GB 2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=24.859 GB
Prediction
zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582bIDdt9cm2jnahrme0ckndkvwzc5zgStatusSucceededSourceWebHardwareL40STotal durationCreatedInput
- seed
- 43
- width
- 960
- height
- 544
- prompt
- A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.
- flow_shift
- 7
- video_length
- 129
- negative_prompt
- Ugly
- num_inference_steps
- 40
- embedded_guidance_scale
- 6
{ "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.", "flow_shift": 7, "video_length": 129, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }
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 zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
const output = await replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", { input: { seed: 43, width: 960, height: 544, prompt: "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.", flow_shift: 7, video_length: 129, negative_prompt: "Ugly", num_inference_steps: 40, embedded_guidance_scale: 6 } } ); // 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 Replicate’s Python client library:pip install replicate
Import the client:import replicate
Run zurk/hunyuan-video-8bit using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.
output = replicate.run( "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", input={ "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.", "flow_shift": 7, "video_length": 129, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } ) print(output)
To learn more, take a look at the guide on getting started with Python.
Run zurk/hunyuan-video-8bit 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": "zurk/hunyuan-video-8bit:392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli\'s whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane\'s sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki\'s enchanting world.", "flow_shift": 7, "video_length": 129, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 } }' \ https://api.replicate.com/v1/predictions
To learn more, take a look at Replicate’s HTTP API reference docs.
Output
{ "completed_at": "2024-12-09T12:52:19.369236Z", "created_at": "2024-12-09T12:31:36.276000Z", "data_removed": false, "error": null, "id": "dt9cm2jnahrme0ckndkvwzc5zg", "input": { "seed": 43, "width": 960, "height": 544, "prompt": "A high-quality, cinematic depiction of a vibrant red seaplane, inspired by Studio Ghibli's whimsical aesthetic, viewed from a top-down perspective as it soars gracefully over a vast, shimmering turquoise sea. The plane's sleek, curving design glides through the golden sunlight, its polished exterior and spinning propeller rendered in stunning clarity. Below, the crystal-clear waters reveal intricate coral reefs, while lush islands with small, high-definition houses and cozy villages nestled among vivid green foliage pass beneath. Cascading waterfalls and meandering paths add charm to the scenery. Puffy white clouds drift lazily across a dreamy, warm-hued sky, amplifying the magical, hand-drawn atmosphere reminiscent of Miyazaki's enchanting world.", "flow_shift": 7, "video_length": 129, "negative_prompt": "Ugly", "num_inference_steps": 40, "embedded_guidance_scale": 6 }, "logs": "2024-12-09 12:36:19,069-INFO-cog_predictor-seed=43\n2024-12-09 12:36:26,516-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 129)\n2024-12-09 12:36:28,928-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 129 frames in 33 latents at 960x544 with 40 inference steps\n 0%| | 0/40 [00:00<?, ?it/s]\n 2%|▎ | 1/40 [00:24<15:48, 24.32s/it]\n 5%|▌ | 2/40 [00:46<14:40, 23.17s/it]\n 8%|▊ | 3/40 [01:09<14:09, 22.97s/it]\n 10%|█ | 4/40 [01:32<13:44, 22.89s/it]\n 12%|█▎ | 5/40 [01:55<13:20, 22.87s/it]\n 15%|█▌ | 6/40 [02:17<12:57, 22.88s/it]\n 18%|█▊ | 7/40 [02:40<12:35, 22.88s/it]\n 20%|██ | 8/40 [03:03<12:12, 22.89s/it]\n 22%|██▎ | 9/40 [03:26<11:50, 22.90s/it]\n 25%|██▌ | 10/40 [03:49<11:27, 22.90s/it]\n 28%|██▊ | 11/40 [04:12<11:04, 22.90s/it]\n 30%|███ | 12/40 [04:35<10:41, 22.90s/it]\n 32%|███▎ | 13/40 [04:58<10:18, 22.89s/it]\n 35%|███▌ | 14/40 [05:21<09:55, 22.89s/it]\n 38%|███▊ | 15/40 [05:43<09:31, 22.88s/it]\n 40%|████ | 16/40 [06:06<09:08, 22.87s/it]\n 42%|████▎ | 17/40 [06:29<08:45, 22.87s/it]\n 45%|████▌ | 18/40 [06:52<08:22, 22.86s/it]\n 48%|████▊ | 19/40 [07:15<07:59, 22.85s/it]\n 50%|█████ | 20/40 [07:38<07:36, 22.85s/it]\n 52%|█████▎ | 21/40 [08:01<07:14, 22.86s/it]\n 55%|█████▌ | 22/40 [08:23<06:51, 22.86s/it]\n 57%|█████▊ | 23/40 [08:46<06:28, 22.85s/it]\n 60%|██████ | 24/40 [09:09<06:05, 22.86s/it]\n 62%|██████▎ | 25/40 [09:32<05:42, 22.85s/it]\n 65%|██████▌ | 26/40 [09:55<05:19, 22.85s/it]\n 68%|██████▊ | 27/40 [10:18<04:56, 22.84s/it]\n 70%|███████ | 28/40 [10:40<04:33, 22.83s/it]\n 72%|███████▎ | 29/40 [11:03<04:11, 22.83s/it]\n 75%|███████▌ | 30/40 [11:26<03:48, 22.83s/it]\n 78%|███████▊ | 31/40 [11:49<03:25, 22.83s/it]\n 80%|████████ | 32/40 [12:12<03:02, 22.82s/it]\n 82%|████████▎ | 33/40 [12:35<02:39, 22.82s/it]\n 85%|████████▌ | 34/40 [12:57<02:16, 22.80s/it]\n 88%|████████▊ | 35/40 [13:20<01:53, 22.79s/it]\n 90%|█████████ | 36/40 [13:43<01:31, 22.78s/it]\n 92%|█████████▎| 37/40 [14:06<01:08, 22.77s/it]\n 95%|█████████▌| 38/40 [14:28<00:45, 22.78s/it]\n 98%|█████████▊| 39/40 [14:51<00:22, 22.77s/it]\n100%|██████████| 40/40 [15:14<00:00, 22.76s/it]\n100%|██████████| 40/40 [15:14<00:00, 22.86s/it]\n2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.993 GB\n2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=22.945 GB\n2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=24.859 GB", "metrics": { "predict_time": 960.297328229, "total_time": 1243.093236 }, "output": "https://replicate.delivery/xezq/0ewFiG7Zmt3ZMy42ofOJHAM0zKRSyVNSGcqUqePGJWnGQTynA/HunyuanVideo_00008.mp4", "started_at": "2024-12-09T12:36:19.071908Z", "status": "succeeded", "urls": { "stream": "https://stream.replicate.com/v1/files/bcwr-pbfqlv57lapek5wimofoj6kib5n2jkl77ryjwppj3sbkxk2ai6ta", "get": "https://api.replicate.com/v1/predictions/dt9cm2jnahrme0ckndkvwzc5zg", "cancel": "https://api.replicate.com/v1/predictions/dt9cm2jnahrme0ckndkvwzc5zg/cancel" }, "version": "392ce81c5d87a6791d8487644b5835983ef1a197d6d77856333997d234ca582b" }
Generated in2024-12-09 12:36:19,069-INFO-cog_predictor-seed=43 2024-12-09 12:36:26,516-INFO-ComfyUI-HunyuanVideoWrapper.utils-Input (height, width, video_length) = (544, 960, 129) 2024-12-09 12:36:28,928-INFO-ComfyUI-HunyuanVideoWrapper.hyvideo.diffusion.pipelines.pipeline_hunyuan_video-Sampling 129 frames in 33 latents at 960x544 with 40 inference steps 0%| | 0/40 [00:00<?, ?it/s] 2%|▎ | 1/40 [00:24<15:48, 24.32s/it] 5%|▌ | 2/40 [00:46<14:40, 23.17s/it] 8%|▊ | 3/40 [01:09<14:09, 22.97s/it] 10%|█ | 4/40 [01:32<13:44, 22.89s/it] 12%|█▎ | 5/40 [01:55<13:20, 22.87s/it] 15%|█▌ | 6/40 [02:17<12:57, 22.88s/it] 18%|█▊ | 7/40 [02:40<12:35, 22.88s/it] 20%|██ | 8/40 [03:03<12:12, 22.89s/it] 22%|██▎ | 9/40 [03:26<11:50, 22.90s/it] 25%|██▌ | 10/40 [03:49<11:27, 22.90s/it] 28%|██▊ | 11/40 [04:12<11:04, 22.90s/it] 30%|███ | 12/40 [04:35<10:41, 22.90s/it] 32%|███▎ | 13/40 [04:58<10:18, 22.89s/it] 35%|███▌ | 14/40 [05:21<09:55, 22.89s/it] 38%|███▊ | 15/40 [05:43<09:31, 22.88s/it] 40%|████ | 16/40 [06:06<09:08, 22.87s/it] 42%|████▎ | 17/40 [06:29<08:45, 22.87s/it] 45%|████▌ | 18/40 [06:52<08:22, 22.86s/it] 48%|████▊ | 19/40 [07:15<07:59, 22.85s/it] 50%|█████ | 20/40 [07:38<07:36, 22.85s/it] 52%|█████▎ | 21/40 [08:01<07:14, 22.86s/it] 55%|█████▌ | 22/40 [08:23<06:51, 22.86s/it] 57%|█████▊ | 23/40 [08:46<06:28, 22.85s/it] 60%|██████ | 24/40 [09:09<06:05, 22.86s/it] 62%|██████▎ | 25/40 [09:32<05:42, 22.85s/it] 65%|██████▌ | 26/40 [09:55<05:19, 22.85s/it] 68%|██████▊ | 27/40 [10:18<04:56, 22.84s/it] 70%|███████ | 28/40 [10:40<04:33, 22.83s/it] 72%|███████▎ | 29/40 [11:03<04:11, 22.83s/it] 75%|███████▌ | 30/40 [11:26<03:48, 22.83s/it] 78%|███████▊ | 31/40 [11:49<03:25, 22.83s/it] 80%|████████ | 32/40 [12:12<03:02, 22.82s/it] 82%|████████▎ | 33/40 [12:35<02:39, 22.82s/it] 85%|████████▌ | 34/40 [12:57<02:16, 22.80s/it] 88%|████████▊ | 35/40 [13:20<01:53, 22.79s/it] 90%|█████████ | 36/40 [13:43<01:31, 22.78s/it] 92%|█████████▎| 37/40 [14:06<01:08, 22.77s/it] 95%|█████████▌| 38/40 [14:28<00:45, 22.78s/it] 98%|█████████▊| 39/40 [14:51<00:22, 22.77s/it] 100%|██████████| 40/40 [15:14<00:00, 22.76s/it] 100%|██████████| 40/40 [15:14<00:00, 22.86s/it] 2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Allocated memory: memory=11.993 GB 2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max allocated memory: max_memory=22.945 GB 2024-12-09 12:51:43,263-INFO-ComfyUI-HunyuanVideoWrapper.utils-Max reserved memory: max_reserved=24.859 GB
Want to make some of these yourself?
Run this model