Pricing for Kubernetes GPU & Cluster Autoscaling
Scale your Kubernetes clusters and GPU workloads
at the right speed and the right cost
Kedify delivers intelligent GPU autoscaling and Kubernetes autoscaling
so your inference, training, and pipelines run efficiently without idle waste.
Starter
Who It’s For
Teams running a few services (APIs, inference endpoints, GPU workloads) in production.
Included
Clusters
Up to 3
Scalers
Support
Business hours
Security
Hardened KEDA build
Marketplace purchase options
Usage
Pricing
Platform package + Usage pricing
Best for
Getting AI APIs production‑ready with intelligent autoscaling, without over‑provisioning.
Price
$800/mo + usage
(billed annually)
Professional
Who It’s For
Platform teams running multiple GPU workloads and clusters, with higher traffic or stricter SLOs.
Included
Clusters
Up to 10
Scalers
+ HTTP scaler (great for inference endpoints and latency SLOs)
Support
Business hours
Security
+ FIPS
Marketplace purchase options
Usage
Pricing
Platform package + Usage pricing
Best for
Consolidating AI APIs, agents, and pipelines across teams with predictable P95s.
Price
$2,000/mo + usage
(billed annually)
Enterprise
Who It’s For
Enterprise‑scale AI platforms with advanced security/compliance and custom scaler needs.
Included
Clusters
Unlimited
Scalers
+ Cluster scaling & AI/GPU
workloads scalers
Support
24/7 on-call with escalation routing
Security
+ FIPS
Airgapped installation
Yes
Marketplace purchase options
Usage
Pricing
Platform package + Usage pricing
Best for
Large‑scale inference, agent fleets, and mission‑critical AI services.
Price
Speak with Sales
Are you an engineer?
Try Kedify for free here.
Usage Pricing
Smart pricing that scales with you
Simple, Transparent, Aligned
- Billed quarterly based on real usage
- No surprises, no lock-in
- Seamless pass-through via your cloud provider
Why it Works
- We charge based on the infrastructure we help you scale; not seats or guesswork.
What’s your annual cloud spend?
Your Estimated Usage Price*:
$6,250 / month
*For a custom quote please talk to sales, your
instance could be different.
Get Your Savings
+ Pricing Estimate
Use our calculator to see your
projected cost based on cluster size.
Pricing is usage- and outcome-based.
No surprise bills.
No overcommitment.
Why teams choose Kedify
Inference & AI APIs:
Scale HTTP/gRPC endpoints to match bursty traffic while holding latency targets.
Agents & RAG Pipelines:
Scale workers on queue depth/lag (Kafka, Redis, RabbitMQ, SQS, and more).
Observability built‑in:
Prometheus/OTel metrics with long‑term, multi‑tenant storage.
Enterprise‑ready:
FIPS builds, SSO via Auth0, marketplace procurement, and air‑gapped options.
Buy Through AWS, GCP or Red Hat Marketplace
Easier procurement. Faster approvals.
Purchase Kedify via AWS, GCP, or Red Hat Marketplaces to count towards cloud spend, bypass vendor delays, and accelerate buying.
Paid Proof of Concept (credited)
Prove impact on your real AI workloads—then roll forward.
Cost:
$5K (credited toward annual contract)
Duration:
30-45 days
Scope:
1 cluster, HTTP scaler, telemetry, dashboard
Includes:
Full onboarding, support, cost tracking, performance metrics
Ready to see it?
We’ll give you a personalized proposal
based on your cluster
footprint and
performance goals.