Changelog #55 - Serverless GPUs in Private Preview, performance improvements on API endpoints, and new 1-click apps

Hello, and welcome to this week’s changelog update! Let’s jump into what’s new:

  1. Serverless GPUs in Private Preview

    We’re excited to share that Serverless GPUs are available for all your AI inference needs directly through the Koyeb platform! We’re starting with 4 Instances: RTX 4000 SFF ADA, L4, V100, and L40S. These GPUs provide up to 48GB of vRAM, 733 TFLOPS and 900GB/s of memory bandwidth to support large models including LLMs and text-to-image models. To access these GPU Instances, join the preview on

  2. Performance improvements on API endpoints

    We’ve improved the performance of some API endpoints. It is now faster to list Koyeb resources like Apps, Services, and Secrets so you can quickly access the current state of your infrastructure.

  3. New one-click apps and tutorials: Directus, Flowise, and Wasp

    We published three new tutorials to deploy Directus, Flowise, and Wasp on Koyeb. Directus is an open data platform that serves as a headless CMS and powerful Backend-as-a-Service (BaaS). Flowise is an open-source AI tool primarily oriented towards building custom LLM workflows and creating AI agents. Wasp is a full-stack web framework built to provide a modern take on a Rails-like development experience.