pip install kaio
from kaio import Client
# Initialize client
client = Client("https://api.kaion5.com")
# Login with your API key
client.login("your-api-key")
# Submit a job
result = client.submit_job(
directory="./tool-calling",
job_name="gemma4-e4b-finetune-tool-calling",
workflow="training",
instance_type="ml.g4dn.xlarge",
entrypoint="finetune.py",
max_runtime_hours=16,
)
Every job is one call to submit_job. Choose the workflow that fits the work, and Kaio runs your code on the instances you ask for, streams its logs and metrics to your dashboard, and hands back what it produced.
Processing jobs take a Python script or a notebook, run it on cloud CPUs or GPUs, and return what it produced. They suit the work around your models: preparing and cleaning data, engineering features, evaluating a model, or scoring a dataset with one.
Kaio packages your folder, installs your requirements.txt on the instance and runs your entrypoint. The job keeps running in the cloud while you carry on working.
result = client.submit_job(
directory="./sevir-nowcast",
job_name="sevir-prep-vil-sequences",
workflow="processing",
instance_type="ml.m5.xlarge",
entrypoint="prep_sequences.py",
)Ask for several instances and each runs your script on its own share of the data. Every instance knows which one it is, so your code can pick its rows.
# submitted with instance_count=4
hosts = os.environ["SM_HOSTS"].split(",")
rank = hosts.index(os.environ["SM_CURRENT_HOST"])
my_rows = dataset[rank::len(hosts)]Point the job at an .ipynb file and every cell runs top to bottom on the instance. The executed notebook comes back with its outputs, charts included.
client.submit_job(
directory="./notebooks",
workflow="processing",
entrypoint="analysis.ipynb",
)Files your code creates or changes are kept as the job's output; your own code isn't copied back. Download them with one call or from the dashboard.
files = client.download_output(
result["unique_job_name"], "./results"
)Training jobs are for fitting models, usually on GPUs. They run your code the same way as processing jobs, and add what long runs need: checkpoints that survive an interruption, and spot capacity at AWS's discount.
The image matches the PyTorch or TensorFlow you use locally, so the training script you tested on your machine is the one that runs.
result = client.submit_job(
directory="./sevir-nowcast",
job_name="convlstm-train-sevir-nowcast",
workflow="training",
instance_type="ml.g4dn.xlarge",
entrypoint="train_convlstm.py",
)Anything your code writes to /opt/ml/checkpoints is saved while the job runs. Load the latest checkpoint at the start, and a restarted job carries on where it stopped.
ckpt = "/opt/ml/checkpoints/model.pt"
if os.path.exists(ckpt):
model.load_state_dict(torch.load(ckpt))
for epoch in range(epochs):
train_one_epoch(model)
torch.save(model.state_dict(), ckpt)Train models at AWS's spot discount, paying only for the time AWS bills. The saving comes from AWS's spare capacity, so a spot job may wait to start or be interrupted, and it picks up again from your checkpoints.
client.submit_job(
directory="./sevir-nowcast",
workflow="training",
training_args={"spot_instances": {
"enabled": True, "max_wait": 7200,
}},
)When the job finishes, the files your code wrote come back packed as model.tar.gz, ready to download alongside the job's logs.
files = client.download_output(
result["unique_job_name"], "./model"
)For scientists, developers, engineers, and python people around the world, Kaio is an MLOps tool that works directly from your IDE, accelerating and enhancing data processing and ML model building workflows while freeing you from cloud compute and infrastructure overhead. Sign up today and focus entirely on speed, experimentation, and discovery.
Sign up now