Access 21+ GPU cloud providers through a single API. Unified billing, automatic failover, zero subscriptions. Pay only for what you use.
Routing your workloads across
Stop juggling multiple cloud accounts. Clearly gives you one gateway to every major GPU provider.
One API to access all providers. Write once, deploy anywhere. No vendor-specific code required.
Automatically routes your jobs to the optimal provider based on cost, availability, and performance.
Spot instance terminated? Your job automatically continues on another provider. Zero downtime.
See exactly what you pay. No hidden fees, no markups. Provider costs plus our small margin.
Load credits once, use them anywhere. No subscription, no commitments. Full control over spending.
GDPR compliant. Data residency options. Run workloads on EU-only infrastructure when you need to.
Get started with a few lines of code. Our API is designed to be intuitive for developers who've used any cloud provider before.
from clearly import Client client = Client() # Launch a GPU instance - routed automatically instance = client.instances.create( gpu_type="A100-80GB", image="pytorch:2.1", spot=True ) # Run your training job instance.run("python train.py") # Auto-failover if spot is preempted print(f"Running on: {instance.provider}")
Start with €100 free credits. No credit card required.
Get Started Free