replicate.com

# Run AI with an API

Run and fine-tune open-source models. Deploy custom models at scale. All with one line of code.

## What You Can Do With Replicate

- Generate images
- Generate text
- Caption images
- Generate music
- Generate speech
- Fine tune models
- Restore images

## Run Open-Source Models

Our community has already published thousands of models that are ready to use in production. You can run these with one line of code.

```python
import replicate

output = replicate.run(
    "black-forest-labs/flux-schnell:f2ab8a5bfe79f02f0789a146cf5e73d2a4ff2684a98c",
    input={
        "prompt": "An astronaut riding a rainbow unicorn, cinematic, dramatic",
        "num_outputs": 1,
        "aspect_ratio": "1:1",
        "output_format": "webp",
        "output_quality": 90
    }
)
print(output)
```

## Fine-tune Models with Your Own Data

You can improve open-source models with your own data to create new models that are better suited to specific tasks. Image models like Flux can generate images of a particular person, object, or style.

Train a model:

```python
import replicate

training = replicate.trainings.create(
    version="ostris/flux-dev-lora-trainer:1296f0ab2d695af5a1b5eeee6e8ec043145b",
    input={
        "input_images": "https://my-domain/my-input-images.zip",
    },
    destination="electricdreams/flux-fine-tuned"
)
print(training)
```

## Deploy Custom Models

You aren't limited to the models on Replicate: you can deploy your own custom models using Cog, our open-source tool for packaging machine learning models.

Define your environment in `cog.yaml`:

```yaml
build:
    gpu: true
    system_packages:
        - "libgl1-mesa-glx"
        - "libglib2.0-0"
    python_version: "3.10"
    python_packages:
        - "torch==1.13.1"
    predict: "predict.py:Predictor"
```

Define predictions in `predict.py`:

```python
from cog import BasePredictor, Input, Path
import torch

class Predictor(BasePredictor):
    def setup(self):
        """Load the model into memory to make running multiple predictions efficient"""
        self.model = torch.load("./weights.pth")

    def predict(self,
               image: Path = Input(description="Grayscale input image")
               ) -> Path:
        """Run a single prediction on the model"""
        processed_image = preprocess(image)
        output = self.model(processed_image)
        return postprocess(output)
```

## Pricing

Pay only for what you use:

- CPU: $0.000100/sec
- Nvidia T4 GPU: $0.000225/sec
- Nvidia A40 GPU: $0.000575/sec
- Nvidia A40 (Large) GPU: $0.000725/sec
- Nvidia A100 (40GB) GPU: $0.001150/sec
- Nvidia A100 (80GB) GPU: $0.001400/sec
- 8x Nvidia A40 (Large) GPU: $0.005800/sec

## Key Features

### Automatic Scaling
If you get a ton of traffic, Replicate scales up automatically to handle the demand. If you don't get any traffic, we scale down to zero and don't charge you a thing.

### Infrastructure Management
Forget about managing infrastructure. We handle API servers, dependencies, model weights, CUDA, GPUs, and batching.

### Monitoring & Logging
Built-in metrics let you monitor model performance, while detailed logs help you debug specific predictions.

## Get Started

With Replicate and tools like Next.js and Vercel, you can wake up with an idea and watch it hit the front page of Hacker News by the time you go to bed.

[Get started with Replicate today →](https://replicate.com)