Koyeb
Serverless platform that deploys code, containers or models with a Git push or CLI call on dedicated GPUs and CPUs in 50+ locations, with autoscaling, scale-to-zero and per-second billing, plus sandboxes for AI agents; per-hour prices for each GPU, from $0.50 an hour, are published.
What it does
- GPU cloud
- Model hosting
- Model inference
How it compares
Sources
- Pricing
-
Per-hour and per-second GPU instance prices, from RTX-4000-SFF-ADA at $0.50 an hour to H200 at $3.00 an hour. Pricing page: "RTX-A6000 vRAM 48GB vRAM 48GB $ 0.75 /hour"
- Description
-
Official documentation · 15 Nov 2026
Product page: "The serverless runtime for AI Inference Fine-tuning AI agents SaaS platforms"
- What it does
-
Official documentation · 15 Nov 2026
Product page: "The serverless runtime for AI Inference Fine-tuning AI agents SaaS platforms"
- Sold within
-
Sold on its own with published plans.
- Name
-
Official documentation · 15 Nov 2026
Name as published on the product page.
- URL
-
Official documentation · 15 Nov 2026
Product page, read 2026-11-15.
- Deployment
-
Official documentation · 15 Nov 2026
Home page: a hosted platform deployed with a Git push or the CLI.
- Interfaces
-
Official documentation · 15 Nov 2026
Home page: a hosted platform deployed with a Git push or the CLI.
Also from Koyeb
-
Koyeb Sandboxes Infrastructure service
Isolated, ephemeral environments on Koyeb for running AI-generated and untrusted code, created through Python and JavaScript SDKs and billed on the compute and storage used.
Something wrong here? Send a correction — quote this product id: koyeb.