What is included in this Hosted.ai update V2.2.1?

This update introduces major improvements across compute provisioning, infrastructure access, automation, and operational visibility. Highlights include: - unified provisioning across GPU services and virtual machines - bare metal GPU instance support - browser-based SSH access - stronger deployment automation - real-time Kubernetes service health monitoring - performance and scalability improvements across core platform services

What does unified provisioning mean in practice?

It means GPU services and virtual machines can now be managed through a more consistent platform. This helps reduce operational complexity and creates a more standardized experience across workload types.

Why is bare metal support important?

Bare metal is valuable for workloads that need direct hardware control, predictable performance, or specialized infrastructure requirements. It also gives customers more flexibility as they scale beyond standard virtualized environments.

What is browser-based SSH access?

Users can now open an SSH session directly from the hosted.ai interface using an embedded browser-based console. This removes the need for separate SSH clients in many common workflows.

Which compute types support browser-based SSH access?

Browser-based SSH access is available across GPU services, virtual machines, and bare metal instances.

How does this help users?

It reduces friction, speeds up access, and makes it easier to work in environments where installing local SSH tools or using additional networking setup is inconvenient.

What automation improvements are included?

hosted.ai now supports stronger deployment-time automation for GPU services, making it easier to apply software stacks and operational configuration consistently during workload rollout.

Can hosted.ai support more advanced containerized and Kubernetes-style workloads now?

Yes. This update expands support for workloads that benefit from more advanced system and service behavior, including scenarios involving nested container and Kubernetes-style environments.

What is the Kubernetes health monitoring dashboard?

It is a real-time service integrity view in the admin experience that provides health visibility, live status awareness, warnings, and audit tracking for Kubernetes services.

Why does Kubernetes health monitoring dashboard matter?

It improves operational visibility, helps teams diagnose issues faster, and reduces dependence on manual checks.

Were there also stability or usability improvements in this update v2.2.1?

Yes. This update includes a number of platform improvements across service creation, editing flows, billing accuracy, scheduler stability, node handling, and email branding behavior.

Will this update change billing behavior?

The update improves billing accuracy and visibility in several areas. For example, it ensures workspace billing reflects usage more accurately and avoids billing certain GPU services before they are active and running.

Who should care most about this release?

This update is especially relevant for: - teams deploying GPU-backed AI services - operators managing mixed compute environments - infrastructure providers offering GPU services - organizations that need more flexibility across virtualized and bare metal infrastructure

How can customers learn more or get a walkthrough?

Customers can contact hosted.ai to schedule a walkthrough of the new capabilities and discuss how these updates fit their infrastructure needs.

News

hosted.ai 2.2.1 update – Infrastructure flexibility with bare metal, browser SSH, and stronger GPU service automation

hosted.ai has rolled out a major platform update v2.2.1 that makes it easier for teams to deploy, manage, and scale AI infrastructure across GPU services, virtual machines, and bare metal.

Let’s get into it!

#BuildingInPublic – the platform story so far

hosted.ai has rolled out a major platform update that makes it easier for teams to deploy, manage, and scale AI infrastructure across GPU services, virtual machines, and bare metal.

This update brings together several improvements that matter directly to operators, platform teams, and AI builders:

better performance and scalability in core platform orchestration
a more consistent provisioning and management experience across GPU services and VMs
new bare metal GPU instance support
browser-based SSH access across compute types
stronger automation for GPU service deployments
real-time Kubernetes service health visibility

The result is a platform that is easier to operate, more flexible for customers, and better suited for modern AI workloads ranging from inference and application hosting to distributed training and Kubernetes-based deployments.

New to hosted·ai? Learn more about our GPU cloud platform or get in touch for a demo.

What’s new in hosted·ai v2.2.1?

1. Unified provisioning across GPU services and VMs

hosted.ai now delivers a more consistent deployment and runtime experience across GPU services and virtual machines.

This means teams can use a more standardized way to provision workloads, reuse service templates more effectively, and manage operational capabilities more consistently across compute environments.

Why is this important?

reduces operational complexity
improves reliability and consistency
makes it easier to scale GPU services across regions and deployment types

Seems interesting? Learn more about our GPU cloud platform or get in touch for a demo.

2. Bare metal GPU instances

hosted.ai now supports bare metal GPU instances as a first-class compute option.

Customers can provision dedicated hardware with lifecycle management, hardware discovery, integrated pricing and billing, and direct access from the hosted.ai platform.

Why it matters:

creates a smoother path for customers moving into higher-performance or specialized environments
supports workloads that require full hardware control
expands deployment flexibility beyond containers and VMs

This is something you are looking for? Learn more about our GPU cloud platform or get in touch for a demo.

3. Browser-based SSH access

Users can now open an SSH session directly from the browser for GPU services, VMs, and bare metal instances.

This removes the need to install external SSH clients or work around restricted environments just to access infrastructure.

Why is this important?

reduces friction for first-time users
speeds up time-to-access
creates a more consistent access experience across compute types

Learn more about our GPU cloud platform or get in touch for a demo.

4. Stronger automation for GPU service deployment

hosted.ai has expanded automation support, so teams can configure GPU service environments more consistently during deployment.

This includes stronger support for software stack customization and operational setup as part of workload provisioning.

Why is this important?

reduces friction for first-time users
speeds up time-to-access
creates a more consistent access experience across compute types