Authenticate

modal token new


```python
import modal

app = modal.App("my-app")

@app.function()
def hello():
    return "Hello from Modal!"

# Run with: modal run script.py

Container Images

# Build image with dependencies
image = (
    modal.Image.debian_slim(python_version="3.12")
    .pip_install("torch", "transformers", "numpy")
)

app = modal.App("ml-app", image=image)

GPU Functions

@app.function(gpu="H100")
def train_model():
    import torch
    assert torch.cuda.is_available()
    # GPU code here

# Available GPUs: T4, L4, A10, A100, L40S, H100, H200, B200
# Multi-GPU: gpu="H100:8"

Web Endpoints

@app.function()
@modal.web_endpoint(method="POST")
def predict(data: dict):
    result = model.predict(data["input"])
    return {"prediction": result}

# Deploy: modal deploy script.py

Scheduled Jobs

@app.function(schedule=modal.Cron("0 2 * * *"))  # Daily at 2 AM
def daily_backup():
    pass

@app.function(schedule=modal.Period(hours=4))  # Every 4 hours
def refresh_cache():
    pass

Autoscaling

@app.function()
def process_item(item_id: int):
    return analyze(item_id)

@app.local_entrypoint()
def main():
    items = range(1000)
    # Automatically parallelized across containers
    results = list(process_item.map(items))

Persistent Storage

volume = modal.Volume.from_name("my-data", create_if_missing=True)

@app.function(volumes={"/data": volume})
def save_results(data):
    with open("/data/results.txt", "w") as f:
        f.write(data)
    volume.commit()  # Persist changes

Secrets Management

@app.function(secrets=[modal.Secret.from_name("huggingface")])
def download_model():
    import os
    token = os.environ["HF_TOKEN"]

ML Model Serving

@app.cls(gpu="L40S")
class Model:
    @modal.enter()
    def load_model(self):
        from transformers import pipeline
        self.pipe = pipeline("text-classification", device="cuda")

    @modal.method()
    def predict(self, text: str):
        return self.pipe(text)

@app.local_entrypoint()
def main():
    model = Model()
    result = model.predict.remote("Modal is great!")

Resource Configuration

@app.function(
    cpu=8.0,              # 8 CPU cores
    memory=32768,         # 32 GiB RAM
    ephemeral_disk=10240, # 10 GiB disk
    timeout=3600          # 1 hour timeout
)
def memory_intensive_task():
    pass

Best Practices

Pin dependencies for reproducible builds
Use appropriate GPU types - L40S for inference, H100 for training
Leverage caching via Volumes for model weights
Use .map() for parallel processing
Import packages inside functions if not available locally
Store secrets securely - never hardcode API keys

vs Alternatives

Platform	Best For
Modal	Serverless GPUs, autoscaling, Python-native
RunPod	GPU rental, long-running jobs
AWS Lambda	CPU workloads, AWS ecosystem
Replicate	Model hosting, simple deployments

Resources

Docs: https://modal.com/docs
Examples: https://github.com/modal-labs/modal-examples

modal

Modal Serverless Cloud Platform

When to Use

Quick Start