Cloudflare Docs
Workers AI
Cloudflare Docs
Workers AI
GitHub icon
Visit Workers AI on GitHub
Set theme to dark (⇧+D)
  1. Products
  2. Workers AI
  3. Models
  4. Image classification

Image classification

ResNet models perform image classification - they take images as input and classify the major object in the image.

​​ Examples


import { Ai } from '@cloudflare/ai'



export interface Env {
  AI: any;

}



export default {
  async fetch(request: Request, env: Env) {
    const res: any = await fetch("https://cataas.com/cat");
    const blob = await res.arrayBuffer();


    const ai = new Ai(env.AI);
    const inputs = {
        image: [...new Uint8Array(blob)],
    };


    const response = await ai.run("@cf/microsoft/resnet-50", inputs);


    return new Response(JSON.stringify({ inputs: { image: [] }, response }));
  }

}


$ curl https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/run/@cf/microsoft/resnet-50 \
    -X POST \
    -H "Authorization: Bearer {API_TOKEN}" \
    --data-binary @orange-llama.png

Example Workers AI response


{
    "inputs": { "image":[] },
    "response": [
        { "label":"PERSIAN CAT" ,"score":0.4071170687675476 },
        { "label":"PEKINESE", "score":0.23444877564907074 },
        { "label":"FEATHER BOA", "score":0.22562485933303833 },
        { "label":"POMERANIAN", "score":0.033316344022750854 },
        { "label":"JAPANESE SPANIEL", "score":0.024184171110391617 }
    ]

}

​​ API schema

The following schema is based on JSON Schema


{
    "task": "image-classification",
    "tsClass": "AiImageClassification",
    "jsonSchema": {
        "input": {
            "type": "object",
            "properties": {
                "image": {
                    "type": "string",
                    "format": "binary"
                }
            },
            "required": ["image"]
        },
        "output": {
            "type": "array",
            "items": {
                "type": "object",
                "properties": {
                    "score": {
                        "type": "number"
                    },
                    "label": {
                        "type": "string"
                    }
                }
            }
        }
    }

}