# CLIP Features - Core.Today AI API

> CLIP ViT-L/14 embeddings for text AND images (163M+ runs) — puts both in the same vector space for cross-modal search, image dedup, and zero-shot classification. 1 credit per run.

- **Provider**: CLIP
- **Model ID**: andreasjansson/clip-features
- **Category**: Image Generation
- **Credits**: 1 per run (batch inputs with newlines)
- **Speed**: Fast
- **Quality**: High

## Features
- Text AND image embeddings in one space
- 163M+ runs — the CLIP standard on Replicate
- Batch multiple inputs per call (newline-separated)
- 1 credit per run

## Use Cases
- Text-to-image search ('find photos of ...')
- Duplicate/near-duplicate image detection
- Zero-shot image classification
- Content-based recommendation

## API Endpoint

Base URL: https://api.core.today/v1
Create Prediction: POST /predictions
Get Status: GET /predictions/{job_id}

## Authentication
Header: X-API-Key: YOUR_API_KEY

## Input Parameters

### Optional
- **inputs**: string (default: a
b) - Newline-separated inputs — text strings or image URLs starting with http(s)://

## Examples

### Text + Image Embeddings
Embed a text query and an image into the same space
```json
{
  "model": "andreasjansson/clip-features",
  "input": {
    "inputs": "a photo of a red sneaker\nhttps://example.com/sneaker.jpg"
  }
}
```

## Response Format
```json
{
  "job_id": "abc123",
  "status": "pending | processing | completed | failed",
  "result": "URL or data (when completed)"
}
```

## Usage Flow
1. POST /predictions with model and input -> receive job_id
2. GET /predictions/{job_id} -> poll until status is completed or failed
3. Result contains output URL(s)

## Tips
- Mix texts and image URLs in one call — each line becomes one embedding
- Cosine-compare a text query embedding against image embeddings for search
- For pure multilingual text retrieval, multilingual-e5-large is stronger

## Documentation
https://replicate.com/andreasjansson/clip-features