cjwbw / shap-e

Generating Conditional 3D Implicit Functions

Cold

Public
15.2K runs
T4
GitHub
Paper
License

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

Run this model in Node.js with one line of code:

npx create-replicate --model=cjwbw/shap-e

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

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 cjwbw/shap-e using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "cjwbw/shap-e:5957069d5c509126a73c7cb68abcddbb985aeefa4d318e7c63ec1352ce6da68c",
  {
    input: {
      prompt: "a shark",
      save_mesh: false,
      batch_size: 1,
      render_mode: "nerf",
      render_size: 128,
      guidance_scale: 15
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

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

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run cjwbw/shap-e using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "cjwbw/shap-e:5957069d5c509126a73c7cb68abcddbb985aeefa4d318e7c63ec1352ce6da68c",
    input={
        "prompt": "a shark",
        "save_mesh": False,
        "batch_size": 1,
        "render_mode": "nerf",
        "render_size": 128,
        "guidance_scale": 15
    }
)
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=<paste-your-token-here>

Find your API token in your account settings.

Run cjwbw/shap-e 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": "cjwbw/shap-e:5957069d5c509126a73c7cb68abcddbb985aeefa4d318e7c63ec1352ce6da68c",
    "input": {
      "prompt": "a shark",
      "save_mesh": false,
      "batch_size": 1,
      "render_mode": "nerf",
      "render_size": 128,
      "guidance_scale": 15
    }
  }' \
  https://api.replicate.com/v1/predictions

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

Output

{
  "completed_at": "2023-05-11T23:30:15.877784Z",
  "created_at": "2023-05-11T23:24:47.854926Z",
  "data_removed": false,
  "error": null,
  "id": "hl67pdznsbagtmfedjekxk3pdy",
  "input": {
    "prompt": "a shark",
    "batch_size": 1,
    "render_mode": "nerf",
    "render_size": 128,
    "guidance_scale": 15
  },
  "logs": "0%|          | 0/64 [00:00<?, ?it/s]\n  2%|▏         | 1/64 [00:00<00:28,  2.20it/s]\n  3%|▎         | 2/64 [00:00<00:23,  2.59it/s]\n  5%|▍         | 3/64 [00:01<00:22,  2.74it/s]\n  6%|▋         | 4/64 [00:01<00:21,  2.79it/s]\n  8%|▊         | 5/64 [00:01<00:20,  2.84it/s]\n  9%|▉         | 6/64 [00:02<00:20,  2.86it/s]\n 11%|█         | 7/64 [00:02<00:19,  2.87it/s]\n 12%|█▎        | 8/64 [00:02<00:19,  2.89it/s]\n 14%|█▍        | 9/64 [00:03<00:18,  2.90it/s]\n 16%|█▌        | 10/64 [00:03<00:18,  2.90it/s]\n 17%|█▋        | 11/64 [00:03<00:18,  2.90it/s]\n 19%|█▉        | 12/64 [00:04<00:17,  2.90it/s]\n 20%|██        | 13/64 [00:04<00:17,  2.90it/s]\n 22%|██▏       | 14/64 [00:04<00:17,  2.90it/s]\n 23%|██▎       | 15/64 [00:05<00:16,  2.91it/s]\n 25%|██▌       | 16/64 [00:05<00:16,  2.90it/s]\n 27%|██▋       | 17/64 [00:05<00:16,  2.90it/s]\n 28%|██▊       | 18/64 [00:06<00:15,  2.90it/s]\n 30%|██▉       | 19/64 [00:06<00:15,  2.90it/s]\n 31%|███▏      | 20/64 [00:06<00:15,  2.90it/s]\n 33%|███▎      | 21/64 [00:07<00:14,  2.90it/s]\n 34%|███▍      | 22/64 [00:07<00:14,  2.90it/s]\n 36%|███▌      | 23/64 [00:08<00:14,  2.90it/s]\n 38%|███▊      | 24/64 [00:08<00:13,  2.90it/s]\n 39%|███▉      | 25/64 [00:08<00:13,  2.90it/s]\n 41%|████      | 26/64 [00:09<00:13,  2.90it/s]\n 42%|████▏     | 27/64 [00:09<00:12,  2.90it/s]\n 44%|████▍     | 28/64 [00:09<00:12,  2.89it/s]\n 45%|████▌     | 29/64 [00:10<00:12,  2.89it/s]\n 47%|████▋     | 30/64 [00:10<00:11,  2.89it/s]\n 48%|████▊     | 31/64 [00:10<00:11,  2.89it/s]\n 50%|█████     | 32/64 [00:11<00:11,  2.89it/s]\n 52%|█████▏    | 33/64 [00:11<00:10,  2.89it/s]\n 53%|█████▎    | 34/64 [00:11<00:10,  2.89it/s]\n 55%|█████▍    | 35/64 [00:12<00:10,  2.89it/s]\n 56%|█████▋    | 36/64 [00:12<00:09,  2.89it/s]\n 58%|█████▊    | 37/64 [00:12<00:09,  2.89it/s]\n 59%|█████▉    | 38/64 [00:13<00:08,  2.90it/s]\n 61%|██████    | 39/64 [00:13<00:08,  2.90it/s]\n 62%|██████▎   | 40/64 [00:13<00:08,  2.89it/s]\n 64%|██████▍   | 41/64 [00:14<00:07,  2.90it/s]\n 66%|██████▌   | 42/64 [00:14<00:07,  2.90it/s]\n 67%|██████▋   | 43/64 [00:14<00:07,  2.90it/s]\n 69%|██████▉   | 44/64 [00:15<00:06,  2.90it/s]\n 70%|███████   | 45/64 [00:15<00:06,  2.90it/s]\n 72%|███████▏  | 46/64 [00:15<00:06,  2.90it/s]\n 73%|███████▎  | 47/64 [00:16<00:05,  2.90it/s]\n 75%|███████▌  | 48/64 [00:16<00:05,  2.89it/s]\n 77%|███████▋  | 49/64 [00:17<00:05,  2.89it/s]\n 78%|███████▊  | 50/64 [00:17<00:04,  2.89it/s]\n 80%|███████▉  | 51/64 [00:17<00:04,  2.89it/s]\n 81%|████████▏ | 52/64 [00:18<00:04,  2.90it/s]\n 83%|████████▎ | 53/64 [00:18<00:03,  2.90it/s]\n 84%|████████▍ | 54/64 [00:18<00:03,  2.89it/s]\n 86%|████████▌ | 55/64 [00:19<00:03,  2.90it/s]\n 88%|████████▊ | 56/64 [00:19<00:02,  2.89it/s]\n 89%|████████▉ | 57/64 [00:19<00:02,  2.89it/s]\n 91%|█████████ | 58/64 [00:20<00:02,  2.89it/s]\n 92%|█████████▏| 59/64 [00:20<00:01,  2.89it/s]\n 94%|█████████▍| 60/64 [00:20<00:01,  2.89it/s]\n 95%|█████████▌| 61/64 [00:21<00:01,  2.89it/s]\n 97%|█████████▋| 62/64 [00:21<00:00,  2.89it/s]\n 98%|█████████▊| 63/64 [00:21<00:00,  2.89it/s]\n100%|██████████| 64/64 [00:22<00:00,  3.39it/s]\n100%|██████████| 64/64 [00:22<00:00,  2.91it/s]",
  "metrics": {
    "predict_time": 58.115524,
    "total_time": 328.022858
  },
  "output": [
    "https://replicate.delivery/pbxt/H0Canvu42iL0B1NCbZBeAjrruuy8PlNsgo1EDTlJpgjDZZdIA/out_%7Bi%7D.gif"
  ],
  "started_at": "2023-05-11T23:29:17.762260Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/hl67pdznsbagtmfedjekxk3pdy",
    "cancel": "https://api.replicate.com/v1/predictions/hl67pdznsbagtmfedjekxk3pdy/cancel"
  },
  "version": "abfc30dc09f51fe27602185f313860c32d501e7a4af6c5a23872eae80e651cb8"
}

Generated in

58.1 seconds

Tweak it Report View full prediction

0%|          | 0/64 [00:00<?, ?it/s]
  2%|▏         | 1/64 [00:00<00:28,  2.20it/s]
  3%|▎         | 2/64 [00:00<00:23,  2.59it/s]
  5%|▍         | 3/64 [00:01<00:22,  2.74it/s]
  6%|▋         | 4/64 [00:01<00:21,  2.79it/s]
  8%|▊         | 5/64 [00:01<00:20,  2.84it/s]
  9%|▉         | 6/64 [00:02<00:20,  2.86it/s]
 11%|█         | 7/64 [00:02<00:19,  2.87it/s]
 12%|█▎        | 8/64 [00:02<00:19,  2.89it/s]
 14%|█▍        | 9/64 [00:03<00:18,  2.90it/s]
 16%|█▌        | 10/64 [00:03<00:18,  2.90it/s]
 17%|█▋        | 11/64 [00:03<00:18,  2.90it/s]
 19%|█▉        | 12/64 [00:04<00:17,  2.90it/s]
 20%|██        | 13/64 [00:04<00:17,  2.90it/s]
 22%|██▏       | 14/64 [00:04<00:17,  2.90it/s]
 23%|██▎       | 15/64 [00:05<00:16,  2.91it/s]
 25%|██▌       | 16/64 [00:05<00:16,  2.90it/s]
 27%|██▋       | 17/64 [00:05<00:16,  2.90it/s]
 28%|██▊       | 18/64 [00:06<00:15,  2.90it/s]
 30%|██▉       | 19/64 [00:06<00:15,  2.90it/s]
 31%|███▏      | 20/64 [00:06<00:15,  2.90it/s]
 33%|███▎      | 21/64 [00:07<00:14,  2.90it/s]
 34%|███▍      | 22/64 [00:07<00:14,  2.90it/s]
 36%|███▌      | 23/64 [00:08<00:14,  2.90it/s]
 38%|███▊      | 24/64 [00:08<00:13,  2.90it/s]
 39%|███▉      | 25/64 [00:08<00:13,  2.90it/s]
 41%|████      | 26/64 [00:09<00:13,  2.90it/s]
 42%|████▏     | 27/64 [00:09<00:12,  2.90it/s]
 44%|████▍     | 28/64 [00:09<00:12,  2.89it/s]
 45%|████▌     | 29/64 [00:10<00:12,  2.89it/s]
 47%|████▋     | 30/64 [00:10<00:11,  2.89it/s]
 48%|████▊     | 31/64 [00:10<00:11,  2.89it/s]
 50%|█████     | 32/64 [00:11<00:11,  2.89it/s]
 52%|█████▏    | 33/64 [00:11<00:10,  2.89it/s]
 53%|█████▎    | 34/64 [00:11<00:10,  2.89it/s]
 55%|█████▍    | 35/64 [00:12<00:10,  2.89it/s]
 56%|█████▋    | 36/64 [00:12<00:09,  2.89it/s]
 58%|█████▊    | 37/64 [00:12<00:09,  2.89it/s]
 59%|█████▉    | 38/64 [00:13<00:08,  2.90it/s]
 61%|██████    | 39/64 [00:13<00:08,  2.90it/s]
 62%|██████▎   | 40/64 [00:13<00:08,  2.89it/s]
 64%|██████▍   | 41/64 [00:14<00:07,  2.90it/s]
 66%|██████▌   | 42/64 [00:14<00:07,  2.90it/s]
 67%|██████▋   | 43/64 [00:14<00:07,  2.90it/s]
 69%|██████▉   | 44/64 [00:15<00:06,  2.90it/s]
 70%|███████   | 45/64 [00:15<00:06,  2.90it/s]
 72%|███████▏  | 46/64 [00:15<00:06,  2.90it/s]
 73%|███████▎  | 47/64 [00:16<00:05,  2.90it/s]
 75%|███████▌  | 48/64 [00:16<00:05,  2.89it/s]
 77%|███████▋  | 49/64 [00:17<00:05,  2.89it/s]
 78%|███████▊  | 50/64 [00:17<00:04,  2.89it/s]
 80%|███████▉  | 51/64 [00:17<00:04,  2.89it/s]
 81%|████████▏ | 52/64 [00:18<00:04,  2.90it/s]
 83%|████████▎ | 53/64 [00:18<00:03,  2.90it/s]
 84%|████████▍ | 54/64 [00:18<00:03,  2.89it/s]
 86%|████████▌ | 55/64 [00:19<00:03,  2.90it/s]
 88%|████████▊ | 56/64 [00:19<00:02,  2.89it/s]
 89%|████████▉ | 57/64 [00:19<00:02,  2.89it/s]
 91%|█████████ | 58/64 [00:20<00:02,  2.89it/s]
 92%|█████████▏| 59/64 [00:20<00:01,  2.89it/s]
 94%|█████████▍| 60/64 [00:20<00:01,  2.89it/s]
 95%|█████████▌| 61/64 [00:21<00:01,  2.89it/s]
 97%|█████████▋| 62/64 [00:21<00:00,  2.89it/s]
 98%|█████████▊| 63/64 [00:21<00:00,  2.89it/s]
100%|██████████| 64/64 [00:22<00:00,  3.39it/s]
100%|██████████| 64/64 [00:22<00:00,  2.91it/s]

This output was created using a different version of the model, cjwbw/shap-e:abfc30dc.

Examples

View more examples

Run time and cost

This model costs approximately $0.093 to run on Replicate, or 10 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia T4 GPU hardware. Predictions typically complete within 7 minutes. The predict time for this model varies significantly based on the inputs.

Readme

Shap-E

This is the official code and model release for Shap-E: Generating Conditional 3D Implicit Functions from https://github.com/openai/shap-e.