Replicate SDK — Hypery Docs

Overview

The official Replicate SDK works with Hypery Hub by configuring the baseUrl.

Installation

npm install replicate

pip install replicate

Setup

import Replicate from 'replicate'
 
const replicate = new Replicate({
  auth: process.env.GATEWAY_API_KEY,
  baseUrl: 'https://hypery.ai/api/v1'
})

import replicate
 
client = replicate.Client(
    api_token=os.getenv('GATEWAY_API_KEY'),
    base_url='https://hypery.ai/api/v1'
)

Run a Model

Generate an image with automatic polling:

const output = await replicate.run(
  'black-forest-labs/flux-schnell',
  {
    input: {
      prompt: 'a cat wearing sunglasses',
      width: 1024,
      height: 1024
    }
  }
)
 
console.log('Generated image:', output[0])

output = client.run(
    'black-forest-labs/flux-schnell',
    input={
        'prompt': 'a cat wearing sunglasses',
        'width': 1024,
        'height': 1024
    }
)
 
print('Generated image:', output[0])

The SDK automatically polls until the prediction completes.

Create Prediction

For more control, create a prediction manually:

const prediction = await replicate.predictions.create({
  model: 'black-forest-labs/flux-schnell',
  input: {
    prompt: 'a sunset over mountains',
    width: 1024,
    height: 1024
  }
})
 
console.log('Prediction ID:', prediction.id)
console.log('Status:', prediction.status)

prediction = client.predictions.create(
    model='black-forest-labs/flux-schnell',
    input={
        'prompt': 'a sunset over mountains',
        'width': 1024,
        'height': 1024
    }
)
 
print('Prediction ID:', prediction.id)
print('Status:', prediction.status)

Get Prediction

Check the status of a prediction:

const prediction = await replicate.predictions.get('abc123')
 
console.log('Status:', prediction.status)
if (prediction.status === 'succeeded') {
  console.log('Output:', prediction.output)
}

prediction = client.predictions.get('abc123')
 
print('Status:', prediction.status)
if prediction.status == 'succeeded':
    print('Output:', prediction.output)

Wait for Prediction

Wait for a prediction to complete:

// Wait for completion (polls automatically)
const prediction = await replicate.wait(prediction.id)
 
console.log('Final output:', prediction.output)

# Wait for completion (polls automatically)
prediction = client.wait(prediction.id)
 
print('Final output:', prediction.output)

Popular Models

Use these model IDs with the Replicate SDK:

Model	Description
`black-forest-labs/flux-schnell`	Fast, high-quality image generation
`black-forest-labs/flux-dev`	Highest quality (slower)
`stability-ai/stable-diffusion-3`	Versatile image generation
`ideogram-ai/ideogram-v2`	Text in images

Browse all models →

Async/Await Pattern

The SDK handles all the complexity:

// Old way (manual polling)
const prediction = await replicate.predictions.create({...})
while (prediction.status !== 'succeeded') {
  await sleep(1000)
  prediction = await replicate.predictions.get(prediction.id)
}
 
// Easy way (SDK handles it)
const output = await replicate.run('model-name', { input: {...} })

Error Handling

try {
  const output = await replicate.run(
    'black-forest-labs/flux-schnell',
    { input: { prompt: 'a sunset' } }
  )
  
  console.log('Success:', output)
} catch (error) {
  if (error.message.includes('credits')) {
    console.error('Out of credits')
  } else {
    console.error('Error:', error.message)
  }
}

try:
    output = client.run(
        'black-forest-labs/flux-schnell',
        input={'prompt': 'a sunset'}
    )
    
    print('Success:', output)
except Exception as e:
    if 'credits' in str(e):
        print('Out of credits')
    else:
        print('Error:', e)

Next Steps

Predictions API

Learn about the Predictions API

Models

Browse Replicate models