News: Koyeb startup program

High-performance Infrastructure
for

Deploy intensive applications across GPUs, CPUs, and Accelerators in minutes - scale in 50+ locations

Trusted by the most ambitious teams

Next-generation cloud experience

No ops, servers, or infrastructure management.

Extreme performanceExtreme performanceAccelerated infrastructure

Accelerated infrastructure

Run all your models and apps on high-performance CPUs, GPUs, and accelerators from AMD, Intel, and Nvidia.
Automatic scalingAutomatic scalingServerless containers

Serverless containers

Deploy production-grade containers with zero configuration — we scale to hundreds of servers and back to zero in seconds.
GlobalGlobalAvailable globally and locally

Available globally and locally

Improve availability and get sub-100ms latency worldwide with over 50 locations. Pick between one and all locations.
Any stackAny stackBuild and deploy anything

Build and deploy anything

Build APIs, distributed systems, or blazing-fast inference endpoints. Deploy your code, containers, or models with a Git push or CLI call.

The serverless runtime for

From dev to high-throughput inference in minutes
Deploy and scale ML models to production without managing infrastructure.
10x faster inference with dedicated performance
Scale to millions of requests with built-in autoscaling on dedicated GPUs.
80% savings compared to hyperscalers
Combine autoscaling with the most efficient GPUs and accelerators on the market.
Autoscaling with sub-200ms cold-start
Seamlessly scale up from zero to hundreds.
Get Started
AI Inference
Extreme performance
Deploy Now
100%
More performance
<250ms
CPU cold start
10
Datacenters
100k
Developers
Build with the languages and frameworks you love, from web apps to inference

Deploy any application seamlessly with native support for popular languages and Docker containers, without any modification.

Deploy now
Fearless development with your team and instant deployment
Fearless development with your team and instant deployment

Experience collaborative development on high-performance cloud infrastructure with a simple Git push. Build together and push to prod with confidence.

Start now
/start building

Deploy in production with one-click apps

From AI models to full-stack apps and databases, start in seconds.

Enterprise-Ready.

Koyeb is a powerful platform with security built-in at all layers.

Any CloudAny Cloud
10 Regions / 3 Continents10 Regions / 3 Continents
Any HardwareAny Hardware
Certifications
Get in touch
/What our customers say

We use GPUs for training and inference, and Koyeb streamlines our workload deployment, optimizing efficiency and simplifying infrastructure management. This allows us to focus on what really matters without worrying about ops and infra.

Samuel Bernard
Samuel Bernard
Founder
View customer stories

Everything you need for production

GPU, NPU, Accelerators, or just CPU
GPU, NPU, Accelerators, or just CPU

Access a wide range of optimized hardware for scale-out workloads, on demand, in seconds.

Deploy now
Instant API Endpoint
Instant API Endpoint

Just hit deploy to provision an API endpoint ready to handle requests in seconds. No waiting, no config.

Get Started
Smart and Fast Autoscaling
Smart and Fast Autoscaling

Get efficient autoscaling to adapt infrastructure to demand, with imperceptible cold start.

Learn more
Zero-Downtime Deployments
Zero-Downtime Deployments

Enjoy built-in continuous deployment with automatic health checks to prevent bad deployment and ensure you’re always up and running.

View docs
Native HTTP/2, WebSocket, and gRPC Support
Native HTTP/2, WebSocket, and gRPC Support

Stream large or partial responses to end-users and accelerate your connections through a global edge network for instant feedback and responsive applications.

Get started
/and much more
Postgres + pgvector

Store, index, and search embeddings with your data at scale using Koyeb's fully managed Serverless Postgres.

Get Started
Ultra-Fast NVME Storage

Store datasets, models, and fine-tune weights on blazing-fast NVME disks offering extremely high write and read throughput for exceptional performance.

Get started
Logs and Instance Access

Troubleshoot and investigate issues easily using real-time logs, or directly connect to your instances.

Get started
Compute costs
RTX-4000-SFF-ADA
VRAM 20GB
(CPU 6 / RAM 44GB)
$0.50 /hr
$0.00014 /sec
L4
VRAM 24GB
(CPU 6 / RAM 32GB)
$0.70 /hr
$0.000194 /sec
RTX-A6000
VRAM 48GB
(CPU 6 / RAM 64GB)
$0.75 /hr
$0.000208 /sec
L40S
VRAM 48GB
(CPU 15 / RAM 64GB)
$1.55 /hr
$0.000430 /sec
A100
VRAM 80GB
(CPU 15 / RAM 180GB)
$2.0 /hr
$0.000555 /sec
View pricing details
Early stage startup?
Get up to $30k in credits to accelerate your go-to-market with high-performance cloud infrastructure.
Apply now
Pay for what you use
Scale as you grow with transparent pricing starting at $0.0022/h. No commitment, no contracts, no hidden costs. Upgrade anytime to unlock features. Get started for free.
View pricing
/Changelog

Last update from the team

Autoscaling, Serverless GPUs, Croissants, and More! The 2024 Recap
Blog
Jan 10, 2025
Autoscaling, Serverless GPUs, Croissants, and More! The 2024 Recap
Autoscaling, scale to zero, serverless GPUs, and more! Last year was full of amazing milestones on our journey to build a next-generation serverless platform to simplify your AI deployments. Check out the recap to see everything we released in 2024.
Read more
Phi-4
One Click
Jan 09, 2025
Phi-4
Deploy Phi-4 on Koyeb high-performance GPU.
Read more
New plans, deploy AI models in one click, and more
Changelog
Dec 20, 2024
New plans, deploy AI models in one click, and more

  • New plans: Pro and Scale
  • Deploy AI models in one click
  • Control panel: Fixed issue with CPU usage
  • Control panel: Bulk delete secrets
Read more
December Recap: Scale to Zero, Serverless GPU Price Drop, and more
Blog
Dec 19, 2024
December Recap: Scale to Zero, Serverless GPU Price Drop, and more
Since we've released so much goodness in the past couple of weeks, we prepared a recap to make sure you don't miss out on any of our serverless news.
Read more

Deploy AI apps to production in minutes

Koyeb is a developer-friendly serverless platform to deploy apps globally. No-ops, servers, or infrastructure management.
All systems operational
© Koyeb