Prediction

Model

firtoz/trellis:4876f2a8

wvnz280mjdrj00cms1qvr6smcg

Status

Succeeded

Source

Web

Hardware

A100 (80GB)

Total duration

3m 38s

Created

10 months ago by ltejedor

Webhook

–

Input

images: https://replicate.delivery/pbxt/MQmDK1SbOQegtAM5EvfsTGWiZcUVoBxUN55nXyIj8j7NjOPj/tmpv04xqg6e.jpg
seed: 0
randomize_seed: true
generate_color: true
generate_normal: false
generate_model: true
save_gaussian_ply: true
return_no_background: false
ss_guidance_strength: 7.5
ss_sampling_steps: 38
slat_guidance_strength: 3
slat_sampling_steps: 12
mesh_simplify: 0.9
texture_size: 2048

{
  "generate_color": true,
  "generate_model": true,
  "generate_normal": false,
  "images": [
    "https://replicate.delivery/pbxt/MQmDK1SbOQegtAM5EvfsTGWiZcUVoBxUN55nXyIj8j7NjOPj/tmpv04xqg6e.jpg"
  ],
  "mesh_simplify": 0.9,
  "randomize_seed": true,
  "return_no_background": false,
  "save_gaussian_ply": true,
  "seed": 0,
  "slat_guidance_strength": 3,
  "slat_sampling_steps": 12,
  "ss_guidance_strength": 7.5,
  "ss_sampling_steps": 38,
  "texture_size": 2048
}

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=r8_Q4b**********************************

This is your API token. Keep it to yourself.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run firtoz/trellis using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "firtoz/trellis:4876f2a8da1c544772dffa32e8889da4a1bab3a1f5c1937bfcfccb99ae347251",
  {
    input: {
      generate_color: true,
      generate_model: true,
      generate_normal: false,
      images: ["https://replicate.delivery/pbxt/MQmDK1SbOQegtAM5EvfsTGWiZcUVoBxUN55nXyIj8j7NjOPj/tmpv04xqg6e.jpg"],
      mesh_simplify: 0.9,
      randomize_seed: true,
      return_no_background: false,
      save_gaussian_ply: true,
      seed: 0,
      slat_guidance_strength: 3,
      slat_sampling_steps: 12,
      ss_guidance_strength: 7.5,
      ss_sampling_steps: 38,
      texture_size: 2048
    }
  }
);

console.log(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

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=r8_Q4b**********************************

This is your API token. Keep it to yourself.

Import the client:

import replicate

Run firtoz/trellis using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "firtoz/trellis:4876f2a8da1c544772dffa32e8889da4a1bab3a1f5c1937bfcfccb99ae347251",
    input={
        "generate_color": True,
        "generate_model": True,
        "generate_normal": False,
        "images": ["https://replicate.delivery/pbxt/MQmDK1SbOQegtAM5EvfsTGWiZcUVoBxUN55nXyIj8j7NjOPj/tmpv04xqg6e.jpg"],
        "mesh_simplify": 0.9,
        "randomize_seed": True,
        "return_no_background": False,
        "save_gaussian_ply": True,
        "seed": 0,
        "slat_guidance_strength": 3,
        "slat_sampling_steps": 12,
        "ss_guidance_strength": 7.5,
        "ss_sampling_steps": 38,
        "texture_size": 2048
    }
)

print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=r8_Q4b**********************************

This is your API token. Keep it to yourself.

Run firtoz/trellis 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": "firtoz/trellis:4876f2a8da1c544772dffa32e8889da4a1bab3a1f5c1937bfcfccb99ae347251",
    "input": {
      "generate_color": true,
      "generate_model": true,
      "generate_normal": false,
      "images": ["https://replicate.delivery/pbxt/MQmDK1SbOQegtAM5EvfsTGWiZcUVoBxUN55nXyIj8j7NjOPj/tmpv04xqg6e.jpg"],
      "mesh_simplify": 0.9,
      "randomize_seed": true,
      "return_no_background": false,
      "save_gaussian_ply": true,
      "seed": 0,
      "slat_guidance_strength": 3,
      "slat_sampling_steps": 12,
      "ss_guidance_strength": 7.5,
      "ss_sampling_steps": 38,
      "texture_size": 2048
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

Object output with 3 properties

model_file(output.glb)

Download output

Use your mouse to zoom and rotate the model

color_video(output_color.mp4)

Download output

gaussian_ply(output_gaussian.ply)

Download output

Preview of file unavailable

Download output

{
  "id": "wvnz280mjdrj00cms1qvr6smcg",
  "model": "firtoz/trellis",
  "version": "4876f2a8da1c544772dffa32e8889da4a1bab3a1f5c1937bfcfccb99ae347251",
  "input": {
    "generate_color": true,
    "generate_model": true,
    "generate_normal": false,
    "images": [
      "https://replicate.delivery/pbxt/MQmDK1SbOQegtAM5EvfsTGWiZcUVoBxUN55nXyIj8j7NjOPj/tmpv04xqg6e.jpg"
    ],
    "mesh_simplify": 0.9,
    "randomize_seed": true,
    "return_no_background": false,
    "save_gaussian_ply": true,
    "seed": 0,
    "slat_guidance_strength": 3,
    "slat_sampling_steps": 12,
    "ss_guidance_strength": 7.5,
    "ss_sampling_steps": 38,
    "texture_size": 2048
  },
  "logs": "INFO:predict:Loading and preprocessing input images...\nINFO:predict:Using randomized seed: 843466713\nINFO:predict:Running TRELLIS pipeline...\nSampling:   0%|          | 0/38 [00:00<?, ?it/s]\nSampling:   3%|▎         | 1/38 [00:00<00:03,  9.60it/s]\nSampling:   8%|▊         | 3/38 [00:00<00:03, 10.72it/s]\nSampling:  13%|█▎        | 5/38 [00:00<00:03, 10.96it/s]\nSampling:  18%|█▊        | 7/38 [00:00<00:02, 11.07it/s]\nSampling:  24%|██▎       | 9/38 [00:00<00:02, 11.14it/s]\nSampling:  29%|██▉       | 11/38 [00:00<00:02, 11.16it/s]\nSampling:  34%|███▍      | 13/38 [00:01<00:02, 11.18it/s]\nSampling:  39%|███▉      | 15/38 [00:01<00:02, 11.20it/s]\nSampling:  45%|████▍     | 17/38 [00:01<00:01, 11.21it/s]\nSampling:  50%|█████     | 19/38 [00:01<00:01, 11.19it/s]\nSampling:  55%|█████▌    | 21/38 [00:01<00:01, 11.20it/s]\nSampling:  61%|██████    | 23/38 [00:02<00:01, 11.20it/s]\nSampling:  66%|██████▌   | 25/38 [00:02<00:01, 11.19it/s]\nSampling:  71%|███████   | 27/38 [00:02<00:00, 11.18it/s]\nSampling:  76%|███████▋  | 29/38 [00:02<00:00, 11.20it/s]\nSampling:  84%|████████▍ | 32/38 [00:02<00:00, 13.95it/s]\nSampling:  92%|█████████▏| 35/38 [00:02<00:00, 16.13it/s]\nSampling: 100%|██████████| 38/38 [00:03<00:00, 17.84it/s]\nSampling: 100%|██████████| 38/38 [00:03<00:00, 12.65it/s]\nSampling:   0%|          | 0/12 [00:00<?, ?it/s]\nSampling:   8%|▊         | 1/12 [00:00<00:03,  3.62it/s]\nSampling:  17%|█▋        | 2/12 [00:00<00:02,  4.48it/s]\nSampling:  25%|██▌       | 3/12 [00:00<00:01,  4.86it/s]\nSampling:  33%|███▎      | 4/12 [00:00<00:01,  5.07it/s]\nSampling:  42%|████▏     | 5/12 [00:01<00:01,  5.19it/s]\nSampling:  50%|█████     | 6/12 [00:01<00:01,  5.27it/s]\nSampling:  58%|█████▊    | 7/12 [00:01<00:00,  5.31it/s]\nSampling:  67%|██████▋   | 8/12 [00:01<00:00,  5.35it/s]\nSampling:  75%|███████▌  | 9/12 [00:01<00:00,  5.37it/s]\nSampling:  83%|████████▎ | 10/12 [00:01<00:00,  5.39it/s]\nSampling: 100%|██████████| 12/12 [00:02<00:00,  7.05it/s]\nSampling: 100%|██████████| 12/12 [00:02<00:00,  5.65it/s]\nINFO:predict:TRELLIS pipeline complete!\nINFO:predict:Available output formats: dict_keys(['mesh', 'gaussian'])\nINFO:predict:Starting video rendering...\nINFO:predict:Generating color video from gaussian output...\nRendering: 0it [00:00, ?it/s]\nRendering: 19it [00:00, 188.74it/s]\nRendering: 44it [00:00, 220.80it/s]\nRendering: 69it [00:00, 232.86it/s]\nRendering: 93it [00:00, 232.47it/s]\nRendering: 117it [00:00, 233.95it/s]\nRendering: 120it [00:00, 229.34it/s]\nINFO:predict:Available gaussian render types: ['color', 'depth']\nINFO:predict:Generated color video successfully\nINFO:predict:Video rendering complete!\nINFO:predict:Generating GLB model...\nINFO- Loaded 15894 vertices and 31772 faces.\n0% done\n5% done\n11% done\n17% done\n23% done\n29% done\n35% done\n41% done\n47% done\n52% done\n58% done\n64% done\n70% done\n76% done\n82% done\n88% done\n94% done\n100% done\nWARNING- Some cuts were necessary to cope with non manifold configuration.\nRendering: 0it [00:00, ?it/s]\nRendering: 11it [00:00, 107.00it/s]\nRendering: 23it [00:00, 112.74it/s]\nRendering: 35it [00:00, 115.17it/s]\nRendering: 47it [00:00, 114.72it/s]\nRendering: 59it [00:00, 115.36it/s]\nRendering: 71it [00:00, 115.98it/s]\nRendering: 83it [00:00, 116.86it/s]\nRendering: 96it [00:00, 118.12it/s]\nRendering: 100it [00:00, 116.01it/s]\nINFO:predict:GLB model generation complete!\nINFO:predict:Saving Gaussian point cloud as PLY...\nINFO:predict:Gaussian PLY file saved successfully!\nINFO:predict:Prediction complete! Returning results...",
  "output": {
    "color_video": "https://replicate.delivery/yhqm/5Mfj6M4OM9RaOScrbwHKzCzeYFVvwpoCD14Trgualu3ftxWoA/output_color.mp4",
    "combined_video": null,
    "gaussian_ply": "https://replicate.delivery/yhqm/xXtsQbix9ZrqHJ0RSmbnHy8BLGDRTOUv9MhdV9kmHOJwN2CF/output_gaussian.ply",
    "model_file": "https://replicate.delivery/yhqm/Qpl4Nz5JL0bwBheNCtYePtph8B4EtHoWoGVkwTnBMdjA3YLUA/output.glb",
    "no_background_images": null,
    "normal_video": null
  },
  "data_removed": false,
  "error": null,
  "source": "web",
  "status": "succeeded",
  "created_at": "2025-02-02T20:51:50.547Z",
  "started_at": "2025-02-02T20:54:30.706267Z",
  "completed_at": "2025-02-02T20:55:28.816645Z",
  "urls": {
    "cancel": "https://api.replicate.com/v1/predictions/wvnz280mjdrj00cms1qvr6smcg/cancel",
    "get": "https://api.replicate.com/v1/predictions/wvnz280mjdrj00cms1qvr6smcg",
    "web": "https://replicate.com/p/wvnz280mjdrj00cms1qvr6smcg"
  },
  "metrics": {
    "predict_time": 58.110378347,
    "total_time": 218.269645
  }
}

Generated in

58.1 seconds

Tweak it Report

INFO:predict:Loading and preprocessing input images... INFO:predict:Using randomized seed: 843466713 INFO:predict:Running TRELLIS pipeline... Sampling: 0%| | 0/38 [00:00<?, ?it/s] Sampling: 3%|▎ | 1/38 [00:00<00:03, 9.60it/s] Sampling: 8%|▊ | 3/38 [00:00<00:03, 10.72it/s] Sampling: 13%|█▎ | 5/38 [00:00<00:03, 10.96it/s] Sampling: 18%|█▊ | 7/38 [00:00<00:02, 11.07it/s] Sampling: 24%|██▎ | 9/38 [00:00<00:02, 11.14it/s] Sampling: 29%|██▉ | 11/38 [00:00<00:02, 11.16it/s] Sampling: 34%|███▍ | 13/38 [00:01<00:02, 11.18it/s] Sampling: 39%|███▉ | 15/38 [00:01<00:02, 11.20it/s] Sampling: 45%|████▍ | 17/38 [00:01<00:01, 11.21it/s] Sampling: 50%|█████ | 19/38 [00:01<00:01, 11.19it/s] Sampling: 55%|█████▌ | 21/38 [00:01<00:01, 11.20it/s] Sampling: 61%|██████ | 23/38 [00:02<00:01, 11.20it/s] Sampling: 66%|██████▌ | 25/38 [00:02<00:01, 11.19it/s] Sampling: 71%|███████ | 27/38 [00:02<00:00, 11.18it/s] Sampling: 76%|███████▋ | 29/38 [00:02<00:00, 11.20it/s] Sampling: 84%|████████▍ | 32/38 [00:02<00:00, 13.95it/s] Sampling: 92%|█████████▏| 35/38 [00:02<00:00, 16.13it/s] Sampling: 100%|██████████| 38/38 [00:03<00:00, 17.84it/s] Sampling: 100%|██████████| 38/38 [00:03<00:00, 12.65it/s] Sampling: 0%| | 0/12 [00:00<?, ?it/s] Sampling: 8%|▊ | 1/12 [00:00<00:03, 3.62it/s] Sampling: 17%|█▋ | 2/12 [00:00<00:02, 4.48it/s] Sampling: 25%|██▌ | 3/12 [00:00<00:01, 4.86it/s] Sampling: 33%|███▎ | 4/12 [00:00<00:01, 5.07it/s] Sampling: 42%|████▏ | 5/12 [00:01<00:01, 5.19it/s] Sampling: 50%|█████ | 6/12 [00:01<00:01, 5.27it/s] Sampling: 58%|█████▊ | 7/12 [00:01<00:00, 5.31it/s] Sampling: 67%|██████▋ | 8/12 [00:01<00:00, 5.35it/s] Sampling: 75%|███████▌ | 9/12 [00:01<00:00, 5.37it/s] Sampling: 83%|████████▎ | 10/12 [00:01<00:00, 5.39it/s] Sampling: 100%|██████████| 12/12 [00:02<00:00, 7.05it/s] Sampling: 100%|██████████| 12/12 [00:02<00:00, 5.65it/s] INFO:predict:TRELLIS pipeline complete! INFO:predict:Available output formats: dict_keys(['mesh', 'gaussian']) INFO:predict:Starting video rendering... INFO:predict:Generating color video from gaussian output... Rendering: 0it [00:00, ?it/s] Rendering: 19it [00:00, 188.74it/s] Rendering: 44it [00:00, 220.80it/s] Rendering: 69it [00:00, 232.86it/s] Rendering: 93it [00:00, 232.47it/s] Rendering: 117it [00:00, 233.95it/s] Rendering: 120it [00:00, 229.34it/s] INFO:predict:Available gaussian render types: ['color', 'depth'] INFO:predict:Generated color video successfully INFO:predict:Video rendering complete! INFO:predict:Generating GLB model... INFO- Loaded 15894 vertices and 31772 faces. 0% done 5% done 11% done 17% done 23% done 29% done 35% done 41% done 47% done 52% done 58% done 64% done 70% done 76% done 82% done 88% done 94% done 100% done WARNING- Some cuts were necessary to cope with non manifold configuration. Rendering: 0it [00:00, ?it/s] Rendering: 11it [00:00, 107.00it/s] Rendering: 23it [00:00, 112.74it/s] Rendering: 35it [00:00, 115.17it/s] Rendering: 47it [00:00, 114.72it/s] Rendering: 59it [00:00, 115.36it/s] Rendering: 71it [00:00, 115.98it/s] Rendering: 83it [00:00, 116.86it/s] Rendering: 96it [00:00, 118.12it/s] Rendering: 100it [00:00, 116.01it/s] INFO:predict:GLB model generation complete! INFO:predict:Saving Gaussian point cloud as PLY... INFO:predict:Gaussian PLY file saved successfully! INFO:predict:Prediction complete! Returning results...