ZenML or ClearML? Which MLOps tool strikes best?

There’s a ton of MLOps tool and I gathered into two open source version that I wanted to try

Hampus Londögård


May 5, 2024

Making it as few words as possible.

Tool Pro Con
ClearML Simple & Everything “fits” Locked into ClearML, i.e. cannot use best tool for the job
ZenML Composable & Extendable Multiple tools to get job done (e.g. MLFlow not visualized in UI)


There’s a lot of similarities, it’s quite easy to get started.

Building Pipelines

They both have the possibility to use decorators which makes the code very simple to read, alas the ClearML way of doing things is not quite as smooth as ZenML.

ZenML builds pipelines and tasks/components in a simpler better way.

Tracking Experiments

To track experiments I believe both solutions got you covered. ClearML’s experiment tracker is quite good and works as you’d expect, while ZenML you decide which tool you want to use (I opted to MLFlow).

ZenML supports: Comet, MLFlow, Neptune, WandB, & Custom. ClearML supports: ClearML.

It’s a draw, ZenML supports “better” trackers BUT ClearML has a native integration which makes things a lot easier.


Both have a simple to use orchestrator. Once again ZenML leans back towards the giants while ClearML uses a built-in native orchestrator that binds everything together.

It’s a draw.


One of the more important parts of a tool is the UI. Here I believe in a way ZenML is strong as they “off-load” each components UI to the component itself, i.e. MLFlow tracing is shown in MLFlow UI.

The UI itself of each tool, i.e. WandB, is much better than ClearML’s offering in my opinion.
But the integration of ClearML as a tool “solve all” is a HUGE timesaver and I think could outweigh using the “better” tooling. Integrating everything from Experiment Comparison to Report Building is an quite amazing feat that I think is worthwhile applauding.


First and foremost, I see both Open Source offering moving more and more towards a SaaS. This is clearly visible by locking certain features in the UI (ZenML, the new UI is beautiful but locked down without your Cloud offering). It’s also shown by supplying additional superb features even when self-hosted. I do understand the need to pay your bills, but it’s sad to see Open Source moving to this either way.

See ZenML comparison (Open Source <> Cloud) and ClearML one.

Sometimes the best option is to opt for the “cloud-native” one, i.e. AWS/Azure/GCP tools. But I love open source… :)

Anyhow, to finalize here’s my judgement:

  • If you prefer to keep your stack as simple as possible: ClearML.
  • If you prefer to keep your stack customized having the best tool for each part: ZenML.

I cannot pick a winner, ZenML enables simpler transition and better tooling all in all, but the full-on integration of ClearML with “everything working together” is quite magical and similar to the cloud-native options (AWS Sagemaker/Azure MLStudio/GCP Vertex).

Find the code for each framework running MNIST: ….

Thanks for this time, Hampus


task = Task.init(
    project_name="MNIST Digit Recognition",
    task_name="Simple NN model with PyTorch Lightning",

class SimpleNN(pl.LightningModule):
    def __init__(self):
        super(SimpleNN, self).__init__()
        self.fc1 = nn.Linear(784, 512)
        self.dropout = nn.Dropout(0.2)
        self.fc2 = nn.Linear(512, 10)

    def forward(self, x):
        x = torch.flatten(x, 1)
        x = self.fc1(x)
        x = torch.relu(x)
        x = self.dropout(x)
        x = self.fc2(x)
        return torch.log_softmax(x, dim=1)

    def training_step(self, batch, batch_idx):
        data, target = batch
        output = self(data)
        loss = nn.functional.cross_entropy(output, target)
        self.log("train_loss", loss)
        return loss

    def test_step(self, batch, batch_idx):
        return self(batch[0])

    def configure_optimizers(self):
        return optim.Adam(self.parameters(), lr=0.001)

params_dictionary = {"epochs": 3}

transform = transforms.Compose(
    [transforms.ToTensor(), transforms.Normalize((0.5,), (0.5,))]

train_dataset = datasets.MNIST(
    root="./data", train=True, transform=transform, download=True
test_dataset = datasets.MNIST(root="./data", train=False, transform=transform)

train_loader = torch.utils.data.DataLoader(
    dataset=train_dataset, batch_size=128, shuffle=True
test_loader = torch.utils.data.DataLoader(
    dataset=test_dataset, batch_size=128, shuffle=False

model = SimpleNN()
trainer = pl.Trainer(max_epochs=params_dictionary["epochs"])
trainer.fit(model, train_loader)


def load_mnist() -> Tuple[
    Annotated[torch.utils.data.DataLoader, "train_loader"],
    Annotated[torch.utils.data.DataLoader, "test_loader"],
    return train_loader, test_loader

def train_model(
    train_loader: torch.utils.data.DataLoader, test_loader: torch.utils.data.DataLoader
    model = SimpleNN()
    trainer = pl.Trainer()
    trainer.fit(model, train_loader)

def train_pipeline():
    train_loader, test_loader = load_mnist()
    train_model(train_loader, test_loader)

if __name__ == "__main__":