smoretalk/clip-interrogator-turbo – Run with an API on Replicate

smoretalk / clip-interrogator-turbo

@pharmapsychotic 's CLIP-Interrogator, but 3x faster and more accurate. Specialized on SDXL.

Warm

Public
2M runs
L40S

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=smoretalk/clip-interrogator-turbo

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";

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

Run smoretalk/clip-interrogator-turbo using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "smoretalk/clip-interrogator-turbo:36a9275d4215df72f67de0daa59929fa24866351405b74b8bcdc1991441aafec",
  {
    input: {
      mode: "best",
      image: "https://replicate.delivery/pbxt/KgRWg4JUfnnszNV78fo1PMRvYxCD9nCgf26Va4RtLxWcuujW/illust-car.png"
    }
  }
);

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=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run smoretalk/clip-interrogator-turbo using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "smoretalk/clip-interrogator-turbo:36a9275d4215df72f67de0daa59929fa24866351405b74b8bcdc1991441aafec",
    input={
        "mode": "best",
        "image": "https://replicate.delivery/pbxt/KgRWg4JUfnnszNV78fo1PMRvYxCD9nCgf26Va4RtLxWcuujW/illust-car.png"
    }
)
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 smoretalk/clip-interrogator-turbo 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": "smoretalk/clip-interrogator-turbo:36a9275d4215df72f67de0daa59929fa24866351405b74b8bcdc1991441aafec",
    "input": {
      "mode": "best",
      "image": "https://replicate.delivery/pbxt/KgRWg4JUfnnszNV78fo1PMRvYxCD9nCgf26Va4RtLxWcuujW/illust-car.png"
    }
  }' \
  https://api.replicate.com/v1/predictions

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

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/smoretalk/clip-interrogator-turbo@sha256:36a9275d4215df72f67de0daa59929fa24866351405b74b8bcdc1991441aafec \
  -i 'mode="best"' \
  -i 'image="https://replicate.delivery/pbxt/KgRWg4JUfnnszNV78fo1PMRvYxCD9nCgf26Va4RtLxWcuujW/illust-car.png"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/smoretalk/clip-interrogator-turbo@sha256:36a9275d4215df72f67de0daa59929fa24866351405b74b8bcdc1991441aafec
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "mode": "best",
      "image": "https://replicate.delivery/pbxt/KgRWg4JUfnnszNV78fo1PMRvYxCD9nCgf26Va4RtLxWcuujW/illust-car.png"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

a painting of a yellow car parked in front of a house with trees in the, by Makoto Shinkai, featured on pixiv, makoto shinkai. high detail, by makoto shinkai, in style of makoto shinkai, makoto shinkai art style, makoto shinkai. —h 2160, anime keyframe

{
  "completed_at": "2024-04-03T16:02:55.865414Z",
  "created_at": "2024-04-03T15:59:47.985000Z",
  "data_removed": false,
  "error": null,
  "id": "58wq98gs25rgg0cemjrbw3kj1w",
  "input": {
    "mode": "best",
    "image": "https://replicate.delivery/pbxt/KgRWg4JUfnnszNV78fo1PMRvYxCD9nCgf26Va4RtLxWcuujW/illust-car.png",
    "style_only": false
  },
  "logs": "Both `max_new_tokens` (=16) and `max_length`(=212) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)\n/root/.pyenv/versions/3.11.9/lib/python3.11/site-packages/transformers/generation/utils.py:1156: UserWarning: Unfeasible length constraints: `min_length` (37) is larger than the maximum possible length (16). Generation will stop at the defined maximum length. You should decrease the minimum length and/or increase the maximum length.\nwarnings.warn(\nFinish generating a caption in 1.91 seconds.\nUnused or unrecognized kwargs: padding.\nUnused or unrecognized kwargs: padding.\nFinish generating an image embedding in 0.23 seconds.\nFinish generating a text embedding in 0.15 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.15 seconds.\nFinish generating a text embedding in 0.15 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.15 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.14 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.13 seconds.\nFinish generating a text embedding in 0.15 seconds.\nFinish generating a text embedding in 0.13 seconds.\nBest prompt is 'a painting of a yellow car parked in front of a house with trees in the, by Makoto Shinkai, featured on pixiv, makoto shinkai. high detail,  by makoto shinkai,  in style of makoto shinkai,  makoto shinkai art style,  makoto shinkai. —h 2160,  anime keyframe'.\nFinish generating a prompt in 8.34 seconds.",
  "metrics": {
    "predict_time": 9.309009,
    "total_time": 187.880414
  },
  "output": "a painting of a yellow car parked in front of a house with trees in the, by Makoto Shinkai, featured on pixiv, makoto shinkai. high detail, by makoto shinkai, in style of makoto shinkai, makoto shinkai art style, makoto shinkai. —h 2160, anime keyframe",
  "started_at": "2024-04-03T16:02:46.556405Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/58wq98gs25rgg0cemjrbw3kj1w",
    "cancel": "https://api.replicate.com/v1/predictions/58wq98gs25rgg0cemjrbw3kj1w/cancel"
  },
  "version": "f66767bbd5c60b841e05dfb72a176d11bf36eca0554c6ea5d6d705cf617bafe8"
}

Generated in

9.3 seconds

Tweak it Share Report

Both `max_new_tokens` (=16) and `max_length`(=212) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
/root/.pyenv/versions/3.11.9/lib/python3.11/site-packages/transformers/generation/utils.py:1156: UserWarning: Unfeasible length constraints: `min_length` (37) is larger than the maximum possible length (16). Generation will stop at the defined maximum length. You should decrease the minimum length and/or increase the maximum length.
warnings.warn(
Finish generating a caption in 1.91 seconds.
Unused or unrecognized kwargs: padding.
Unused or unrecognized kwargs: padding.
Finish generating an image embedding in 0.23 seconds.
Finish generating a text embedding in 0.15 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.15 seconds.
Finish generating a text embedding in 0.15 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.15 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.14 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.13 seconds.
Finish generating a text embedding in 0.15 seconds.
Finish generating a text embedding in 0.13 seconds.
Best prompt is 'a painting of a yellow car parked in front of a house with trees in the, by Makoto Shinkai, featured on pixiv, makoto shinkai. high detail,  by makoto shinkai,  in style of makoto shinkai,  makoto shinkai art style,  makoto shinkai. —h 2160,  anime keyframe'.
Finish generating a prompt in 8.34 seconds.

This output was created using a different version of the model, smoretalk/clip-interrogator-turbo:f66767bb.

Run time and cost

This model costs approximately $0.00098 to run on Replicate, or 1020 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 L40S GPU hardware. Predictions typically complete within 1 seconds.

Readme

@pharmapsychotic ‘s CLIP-Interrogator, but fully-optimized version. You can get more exact prompt from an image with 3x faster speed. To create cool reference-based image, visit https://flamel.app/