> ## Documentation Index
> Fetch the complete documentation index at: https://docs.nebius.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Nebius AI Cloud changelog
This article lists what has been added, updated, fixed or removed in Nebius AI Cloud services.
* *Added:* You can now [access instance metadata from within a Compute virtual machine](/compute/virtual-machines/instance-metadata) (VM) through the HTTP instance metadata service (IMDS). Retrieve instance details, labels, network and user data, service account tokens and live maintenance events without calling the Nebius AI Cloud API from outside the VM.
* *Updated:* The legacy `/mnt/cloud-metadata` mount on older VM images is deprecated and will be removed on September 30, 2026. Use the [HTTP instance metadata service](/compute/virtual-machines/instance-metadata) instead.
* *Updated:* The `ubuntu22.04-cuda12` and `ubuntu22.04-driverless` [boot disk image families](/compute/storage/boot-disk-images) will be deprecated on June 1, 2026. Existing disks remain available, but you can no longer create new disks from these families after that date. Migrate to Ubuntu 24.04 images before the deadline.
* *Updated:* When [setting up GPUs in Managed Kubernetes](/kubernetes/gpu/set-up), the `cuda12.4` driver preset (Ubuntu 22.04) is no longer available. Use `cuda12.8` or `cuda13.0` on Ubuntu 24.04.
* *Added:* You can now [create a Compute virtual machine and its disks together](/compute/virtual-machines/manage#creating-a-vm-and-disks-together) in a single CLI command. These *managed disks* share the VM's lifecycle and are deleted automatically with it, which makes them a good fit for ephemeral workloads such as training jobs or inference clusters.
* *Added:* You can now [tune TCP window sizes on Linux virtual machines](/compute/virtual-machines/tcp-window-tuning) to improve per-flow throughput on high-bandwidth, high-latency connections.
* *Updated:* You can now set the [block size](/compute/storage/manage#volume-parameters) for a disk or shared filesystem when creating it in the web console, in addition to the CLI and Terraform provider.
* *Deprecated:* The `nebius storage v1alpha1 transfer` CLI commands and the `nebius_storage_v1alpha1_transfer` Terraform resource are deprecated and supported until 2026-06-01. Use the v1 data transfer API instead; existing transfers remain accessible through the v1 API. See the [CLI release notes](/cli/release-notes) and [Terraform provider release notes](/terraform-provider/reference/release-notes).
* *Added:* The [NCCL Inspector](/slurm-soperator/monitoring/nccl-inspector) by NVIDIA® is now available on demand for Soperator clusters in Nebius AI Cloud. When enabled for a job, it collects per-communicator, per-collective performance metrics from NCCL operations and sends them to the Grafana® installation included with your cluster.
* *Added:* Soperator clusters now include built-in [health checks and an auto-healing system](/slurm-soperator/worker-nodes-health) for worker nodes. Critical GPU, RAM and network checks run on a schedule, and broken nodes are isolated and replaced automatically.
* *Added:* You can now configure single sign-on with [Okta](/iam/federations/configure-sso-okta).
* *Added:* Compute virtual machines now expose [maintenance reason codes](/compute/virtual-machines/maintenance-reasons) that explain why a maintenance event was triggered.
* *Added:* You can now [retrieve the public IPv4 CIDR blocks](/vpc/addressing/public-address-ranges) available for allocations in a project. The available ranges depend on the project region.
* *Added:* You can now collect diagnostic logs from [Compute virtual machines](/compute/virtual-machines/logs) and from [Managed Kubernetes clusters](/kubernetes/troubleshooting/logs), including GPU, InfiniBand™ and node-level logs for troubleshooting.
* *Added:* A new [all-reduce NCCL performance test](/slurm-soperator/jobs/examples/nccl-all-reduce) is now included in Soperator. Use the pre-built single-node and multi-node versions to validate collective communication between GPUs over NVLink and InfiniBand and assess network performance.
* *Updated:* SkyPilot as a standalone application is now available in the `uk-south1` [region](/overview/regions#availability-of-services-per-region).
* *Updated:* The [Observability Overview](/observability) page in the web console is now the new entry point for monitoring, providing information on firing alerts, key resource metrics and recent logs in a single view.
* *Added:* Compute disks and shared filesystems now support [deletion protection](/compute/storage/manage#deletion-protection). Enable it to prevent accidental deletion in automated environments such as CI/CD pipelines or Terraform configurations.
* *Added:* The Nebius AI Cloud CLI now supports an [interactive mode](/cli/interactive). To enable it, add `-i` or `--interactive` to a command. The CLI prompts for parameters step by step, validates each value and shows the resulting configuration before you apply it.
* *Added:* The [log export to Object Storage](/observability/logs/export) now supports the Nebius AI Cloud CLI.
* *Added:* You can now get a [list of public boot disk images](/compute/storage/boot-disk-images#how-to-get-a-list-of-available-public-images) available in a given region. The list includes image families, recommended platforms and unsupported platforms for each image.
* *Added:* When Prometheus takes too long to retrieve metrics of large clusters, [shard and split their scraping jobs](/observability/metrics/prometheus#how-to-shard-large-scraping-jobs).
* *Updated:* The AWS CLI is now documented as a supported interface for [configuring lifecycle rules](/object-storage/objects/configure-lifecycle-rules) in Object Storage, alongside the Nebius AI Cloud interfaces.
* *Added:* The [capacity advisor](/compute/virtual-machines/capacity-advisor) is now available in the web console. It shows GPU capacity availability across regions, platforms and presets so you can check where you can launch virtual machines before creating them.
* *Added:* You can now [monitor data plane events](/audit-logs/resources/quotas-limits#monitoring-data-plane-events) for Audit Logs in the web console. Graphs show processed events, quota usage and dropped events so you can detect when Object Storage activity approaches the quota.
* *Updated:* You can use the web console to create, configure and delete [Serverless AI endpoints](/serverless/endpoints/manage).
* *Updated:* You can manage [authorized keys for service accounts](/iam/service-accounts/authorized-keys) from the web console and the Terraform provider, in addition to the CLI.
* *Updated:* The [SkyPilot integration tutorial](/3p-integrations/skypilot) now includes instructions to use the Managed SkyPilot API Server, a standalone application available in the Nebius AI Cloud console.
* *Updated:* The [Nebius Service Health Status](https://status.nebius.com) page now displays all timestamps in UTC (GMT+00:00).
* *Added:* [Nebius Echo](/overview/nebius-echo), an AI assistant, is now available in the web console. It can explain services and features, inspect project resources and perform actions like creating or stopping virtual machines.
* *Added:* You can now view and [reschedule maintenance operations](/mlflow/maintenance) for Managed Service for MLflow clusters. The same capability is available for [Managed Service for PostgreSQL](/postgresql/maintenance).
* *Added:* A new [compute.instance-power-operator](/iam/authorization/roles#compute) IAM role lets you start and stop Compute virtual machines without granting broader Compute permissions.
* *Added:* You can now [query and stream logs directly from the Nebius AI Cloud CLI](/observability/logs/nebius-cli) using LogQL, without configuring external tools.
* *Updated:* [Object Storage monitoring](/object-storage/monitoring) now includes additional bucket metrics: space by storage class, object counts by storage class, API errors and transition lifecycle failures.
* *Updated:* Three new ML libraries — AutoKeras, auto-sklearn and TPOT — are now listed on the [third-party compatibility](/3p-integrations/compatibility) page.
* *Added:* The [NVIDIA® RTX PRO™ 6000](/compute/virtual-machines/types) virtual machine platform is now available in the `us-central1` region. It features 96 GB of GDDR7 memory per GPU and comes in single-GPU and eight-GPU presets.
* *Added:* Serverless AI CLI commands now support [MysteryBox secrets](/cli/release-notes). You can pass secrets as environment variables, registry credentials or tokens when creating endpoints and jobs, without exposing sensitive values in plain text.
* *Updated:* You can now manage [capacity reservations](/compute/virtual-machines/reservations) for Compute virtual machines and [Managed Kubernetes node groups](/kubernetes/node-groups/reservations) in the web console. Previously, this required the CLI or Terraform provider.
* *Updated:* You can now [export logs](/observability/logs/export) directly from individual resource pages in the web console, in addition to the central Observability Logs page.
* *Updated:* The default retention period for [logs in Observability services](/observability/resources/quotas-limits#logs-constraints) is now 14 days.
* *Added:* [Object versioning](/object-storage/buckets/versioning) is available in Object Storage. When enabled, each update to an object creates a new version, which protects data from loss and allows you to restore previous versions.
* *Added:* You can now create [custom boot disk images](/compute/storage/custom-disk-images) to tailor images to your needs and create multiple virtual machines with custom disk architecture.
* *Added:* You can also create custom boot disk images by using the [Nebius Packer plug-in](/compute/storage/packer). Use it to pre-install packages to a disk, prepare versioned images and standardize environments across teams or workloads.
* *Added:* You can now use the [Nebius Terraform provider with Pulumi](/terraform-provider/pulumi) to declare your Nebius AI Cloud resources in the preferred programming language instead of using Terraform configuration.
* *Added:* You can now [export logs](/observability/logs/export) from Observability to an Object Storage bucket in JSON or Parquet format for long-term retention or analysis in external tools.
* *Updated:* You can now [attach and detach disks](/compute/storage/use#how-to-attach-volumes-to-vms) on running Compute virtual machines without stopping them first.
* *Added:* You can now [view logs for Managed Kubernetes clusters](/kubernetes/logs) in the web console, including optional audit logs and control plane component logs collected by default.
* *Added:* You can now [fine-tune a large language model with Serverless AI jobs](/serverless/tutorials/fine-tuning) by using Axolotl and LoRA, with Object Storage for checkpoints and adapter weights.
* *Added:* The NVIDIA B200 platform now includes a public [single-GPU preset](/compute/virtual-machines/types) (`1gpu-20vcpu-224gb`) in addition to the eight-GPU configuration.
* *Updated:* Managed Kubernetes now uses [Kubernetes 1.33](/kubernetes/versions) for new clusters and node groups. Older supported versions remain available for existing deployments, but are no longer recommended.
* *Updated:* [Preemptible virtual machines](/compute/virtual-machines/preemptible) are supported on all [GPU VM platforms](/compute/virtual-machines/types). [Compute pricing](/compute/resources/pricing) now lists Israeli shekel (ILS) rates for preemptible VMs where that billing currency applies.
* *Updated:* You can now query logs only within the [Logs constraints](/observability/resources/quotas-limits#logs-constraints) time range, from three days in the past to one day ahead of the current time.
* *Added:* A new legal entity, [Nebius Israel Ltd.](/legal/dpa-il), is launched. Companies from Israel can now only pay by [bank transfers](/signup-billing/payments/bank-transfers) in Israeli shekels (ILS, ₪).
* *Added:* You can now configure single sign-on by using [Keycloak](/iam/federations/configure-sso-keycloak).
* *Updated:* GPU worker nodes in [Managed Service for Soperator](/slurm-soperator/deploy/overview#managed-service-for-soperator) are only available if you have capacity block groups that reserve GPUs.
* *Added:* Virtual machines now support [secondary private IP addresses](./compute/virtual-machines/network#how-to-assign-a-secondary-private-ip-address-to-a-vm) that you can use as a backup option in case of incidents.
* *Added:* [Nebius VPN Gateway](./vpc/nebius-vpn-gateway), an open-source, VM-based IPsec gateway provided by Nebius Professional Services, is now available. You can deploy it to establish site-to-site VPN connectivity with external networks in other cloud providers or on-premises data centers.
* *Added:* Nebius AI Cloud now calculates [taxes](./signup-billing/payments/taxes#united-states) for the residents of Rhode Island in accordance with the United States tax law.
* *Updated:* For [viewing ingested traces from Managed Kubernetes clusters](./observability/traces/grafana) in Grafana®, the connection URL is now `https://read.tracing.api.nebius.cloud/projects//tempo` instead of `https://read.tracing.api.nebius.cloud/tempo/`.
* *Added:* [Managed Service for Soperator](./slurm-soperator/deploy/overview#nebius-ai-cloud-managed-solutions) is now generally available. You can create and manage Soperator clusters with just a few clicks.
* *Added:* You can now [collect journald logs from Compute virtual machines](./observability/logs/journald) and forward these logs to Logging.
* *Updated:* [Audit event export](./audit-logs/events/export) to Object Storage buckets is only supported in the `eu-north1` region.
* *Updated:* [Specifications](./compute/storage/types#filesystem-specifications) and [performance numbers](./compute/storage/types#filesystems-performance) for shared filesystems are actualized.
* *Added:* [Security groups](/vpc/security-groups/overview) are now available for Virtual Networks. You can use them as a firewall for your Compute virtual machines and control ingress and egress traffic with [security rules](/vpc/security-groups/manage).
* *Added:* New [Object Storage roles](/iam/authorization/roles) provide more granular access to buckets and objects. See the list of all [actions supported by Object Storage roles](/object-storage/supported-actions).
* *Added:* You can now specify a [service account when creating a Compute virtual machine](/compute/virtual-machines/manage) so that the VM performs actions on your behalf. This setting is available in the [Additional](/compute/virtual-machines/params#additional) section of the VM configuration.
* *Added:* You can now use [CLI exit codes](/cli/exit-codes) to interpret command failures and to automate scripts that depend on command outcomes.
* *Added:* You can now use [Mountpoint for Amazon S3](/object-storage/interfaces/mountpoint-s3) with Object Storage. Mount buckets as local filesystems on your machines or on Compute virtual machines and use the recommendations for use cases, limitations and performance.
* *Added:* You can now [deploy applications on containers over virtual machines](/compute/virtual-machines/applications). This option provides maximum performance tuning and direct control over infrastructure. See all available [deployment options](/applications/types).
* *Added:* [Network SSD NRD and Network SSD IO M3](./compute/storage/types#disk-types) boot disks now support [encryption](./compute/storage/types#encryption-of-disks), allowing you to securely store personal and other sensitive data, while reducing the risk of unauthorized access.
* *Updated:* [Applications](./applications) now have an updated interface that offers improved navigation and filtering, as well as unified product cards with multiple installation options in one place.
* *Updated:* Nebius AI Cloud may [automatically reduce quotas](/overview/quotas#automatic-quota-reduction) when usage stays low for a prolonged period. Users receive an email notification and have a grace period before the change applies.
* *Added:* In Serverless AI, you can now [debug failed jobs](/serverless/jobs/failure) by keeping the job's VM alive to connect via SSH.
* *Added:* [Single sign-on with JumpCloud](/iam/federations/configure-sso-jumpcloud) is now available. Users can sign in to Nebius AI Cloud with their JumpCloud credentials instead of a separate password.
* *Added:* You can now [export audit events to an Object Storage bucket](/audit-logs/events/export) for a specific time range. Use exported data for long-term retention, data portability or analysis in external tools.
* *Added:* You can now [collect traces](/observability/traces/ingest) with Nebius Observability Agent for Kubernetes and [view traces in Grafana](/observability/traces/grafana). Send traces from cluster applications via OTLP and explore them in Grafana with Tempo.
* *Updated:* We now recommend [GPU boot disk images](/compute/storage/boot-disk-images#images-for-gpu-vms) with Ubuntu 24.04 and CUDA 12 or CUDA 13.0. The `ubuntu22.04-cuda12` image is not recommended and will be deprecated.
* *Updated:* When [setting up GPUs in Managed Kubernetes](/kubernetes/gpu/set-up), you can now choose between driver presets `cuda12.8`, `cuda13.0` and `cuda12.4`. You can also [change the driver preset](/kubernetes/gpu/set-up#how-to-change-the-driver-preset) by updating the node group when you need a different driver or CUDA version.
* *Added:* [Serverless AI jobs and endpoints](/serverless/overview) are now available. You can use them to run AI workloads without any need to create and maintain infrastructure for them.
* *Added:* [Taxes](/signup-billing/payments/taxes) are now included in invoices for individuals from India, Singapore, the United Arab Emirates and the United Kingdom.
* *Updated:* You can now use a Microsoft account to [log in to Nebius AI Cloud](/iam/log-in).
* *Added:* Object Storage now supports [logs](/object-storage/logs) of two types: control plane logs and data plane logs. By using them, you can get detailed information about actions on buckets and objects.
* *Added:* You can now use [capacity block groups](/overview/limits/capacity-block-groups) to [reserve GPUs](/compute/virtual-machines/reservations) for virtual machines. Capacity block groups and reservations ensure that you always have the computing resources necessary for your workloads and applications.
* *Added:* The [Managed Service for MLflow](/mlflow) is now available in the `me-west1` and `uk-south1` [regions](/overview/regions).
* *Added:* The [Managed Service for PostgreSQL](/postgresql) is now available in the `uk-south1` [region](/overview/regions).
* *Added:* Object Storage now supports the [Force storage class](/object-storage/buckets/manage#bucket-parameters) parameter for buckets. It overrides a storage class that you specify when you upload an object to a bucket.
* *Added:* You can now set up [billing data export](/signup-billing/usage/export) and automatically get usage data for cost analysis.
* *Updated:* The `me-west1` region is now [public](/overview/regions#public-regions). Services available in this region are listed in [Regions](/overview/regions#availability-of-services-per-region).
* *Updated:* The default [Compute quotas](/compute/resources/quotas-limits) on the numbers of NVIDIA B200 and NVIDIA H200 GPUs in the `us-central1` region per tenant are now zero, down from 32 and 16, respectively. This means that in tenants created after December 1, you need to [request a quota raise](/overview/quotas#change-quotas) before creating VMs with any number of these GPUs in `us-central1` projects. The change does not affect existing tenants.
* *Added:* Two new [virtual machine platforms](/compute/virtual-machines/types) are now available for Compute virtual machines, Managed Kubernetes nodes and other resources:
* NVIDIA® B300 NVLink with Intel Granite Rapids (ID `gpu-b300-sxm`) is now available in the [private region](/overview/regions) `uk-south1`. It introduces NVIDIA B300 GPUs to Nebius AI Cloud.
* NVIDIA® B200 NVLink with Intel Emerald Rapids (ID `gpu-b200-sxm-a`) is now available in the private region `me-west1`. This is the second platform with NVIDIA B200 GPUs; a platform with the same name (ID `gpu-b200-sxm`) is available in the public region `us-central1`.
* *Added:* When creating a [support ticket](/overview/support) in the [support center](https://console.nebius.com/support/create-ticket), you can now choose the topic of the ticket (technical, billing or digital rights). Available priority levels for technical tickets have also changed: the standard level is renamed medium, and new high and low levels are added to the existing levels. This helps us process your requests faster and more efficiently.
* *Added:* Four [NVIDIA NIM](https://www.nvidia.com/en-us/ai-data-science/products/nim-microservices/) microservices are now available as [standalone applications](/applications/standalone/nvidia-nim/index). These microservices provide models from the NVIDIA BioNeMo Framework for computational drug discovery: [Boltz-2](https://build.nvidia.com/mit/boltz2/modelcard), [Evo 2](https://build.nvidia.com/arc/evo2-40b/modelcard), [GenMol](https://build.nvidia.com/nvidia/genmol-generate/modelcard) and [MolMIM](https://build.nvidia.com/nvidia/molmim-generate/modelcard).
* *Added:* The [NVIDIA Blueprint for generative virtual screening](https://github.com/NVIDIA-BioNeMo-blueprints/generative-virtual-screening) is now available as an [application for Managed Kubernetes](/applications/kubernetes/nvidia-blueprint-virtual-screening/index). The blueprint provides an AI-driven pipeline to design optimized small molecules efficiently.
* *Added:* You can now [delete empty projects](/iam/manage-projects#how-to-delete-a-project) that you no longer need, helping you keep your Nebius AI Cloud organized.
* *Updated:* You can now only delay a [maintenance event for Compute virtual machines](/compute/virtual-machines/maintenance) two times, up to seven days each time.
* *Added:* For residents of Texas and Washington, US, Nebius now [calculates taxes](/signup-billing/payments/taxes) in accordance with United States tax law and includes them in invoices.
* *Updated:* The default [Compute quota](/compute/resources/quotas-limits) on the number of NVIDIA H200 GPUs in the `us-central1` region per tenant is now 16, down from 32. The change does not affect existing tenants.
* *Updated:* When you [manually install the NVIDIA GPU Operator](/kubernetes/gpu/set-up#how-to-install-the-drivers-and-components-on-existing-node-groups) on your InfiniBand-interconnected Managed Kubernetes nodes, GPUDirect RDMA is now enabled by default.
* *Added:* Monitoring dashboards for Managed PostgreSQL clusters now include two new [connection pooler metrics](/postgresql/monitoring#connection-pooler), **Pooler clients active** and **Pooler clients waiting**.
* *Added:* Managed Kubernetes now supports Kubernetes 1.32. This version is selected for new clusters and node groups by default. See the [overview of Kubernetes versions](/kubernetes/versions).
* *Added:* The list of [roles](/iam/authorization/roles) has been extended. In addition to general roles (`admin`, `editor`, `viewer`, `auditor`), you can now assign new service roles to [custom groups](/iam/authorization/groups/index#custom-groups), allowing for more granular access to your resources. The new roles control access to Object Storage and MysteryBox resources, as well as Data Subject Requests to support.
* *Updated:* Starting October 22, usage of virtual machines and applications that use NVIDIA B200, H200 and H100 GPUs is recorded under new billing items that cover GPUs, vCPUs and RAM together rather than separately. Usage from October 1–21 is recalculated retroactively. This makes monitoring your usage simpler and does not affect your costs. [Learn more](/signup-billing/billing-changes-oct-2025).
* *Added:* You can now create [preemptible virtual machines](/compute/virtual-machines/preemptible) with NVIDIA B200 GPUs in the `us-central1` region.
* *Added:* You can now [create your own tenants](/iam/create-tenants) for dedicated billing and more resource isolation.
* *Removed:* New Managed MLflow clusters no longer support the `small` size. See the guide to [creating clusters](/mlflow/clusters/manage).
* *Updated:* Automatic security updates are now disabled by default in [boot disk images](/compute/storage/boot-disk-images) for non-GPU virtual machines in Compute, in addition to GPU VMs, to prevent errors in running workloads. You can [enable them back](/compute/storage/automatic-updates).
* *Updated:* Computing resources of all Compute virtual machines are now eligible for [committed usage discounts](/signup-billing/billing-models/committed-usage), as opposed to only VMs with NVIDIA H100 GPUs.
* *Added:* The [MysteryBox](/mysterybox) service is launched, allowing you to store sensitive data in an encrypted form to reuse them in your scripts, configuration files or applications.
* *Added:* You can now create [routing tables and routes](/vpc/routing/overview) to customize the routing of egress traffic and implement advanced networking scenarios, such as [custom NAT gateways](/vpc/routing/custom-nat-gateway) or segmented architectures.
* *Added:* Your [committed usage discounts](/signup-billing/billing-models/committed-usage) are now grouped in orders that correspond to the agreement addendums that you sign.
* *Updated:* Automatic security updates are now disabled by default in [boot disk images](/compute/storage/boot-disk-images) for GPU virtual machines in Compute to prevent errors in running workloads. You can [enable them back](/compute/storage/automatic-updates).
* *Updated:* Default [Compute quotas](/compute/resources/quotas-limits) for GPU and non-GPU virtual machines are now separate, allowing new customers to create more VMs overall and control the remaining capacity in their projects more precisely.
* *Added:* You can now create Compute virtual machines with container images deployed on them: [containers over VMs](/compute/virtual-machines/containers). You can use images provided by Nebius AI Cloud or custom Docker images from public registries.
* *Added:* Two new [extensions for PostgreSQL](/postgresql/databases/extensions) are now available on all Managed Service for PostgreSQL clusters: [RUM](https://github.com/postgrespro/rum) offers enhanced full-text search, and [pg\_repack](https://github.com/reorg/pg_repack) allows you to remove bloat from tables and indexes.
* *Updated:* If you pay for Nebius AI Cloud services [by bank card](/signup-billing/payments/card), we can now suspend your account after one unsuccessful attempt to charge your card rather than after multiple attempts. Check regularly that your payment method is up to date.
* *Added:* Managed Kubernetes now performs [health checks](/kubernetes/maintenance/health-checks) on nodes, and automatically recovers nodes that are not ready or have issues with the boot disk or GPUs. You can [enable and disable](/kubernetes/maintenance/enable-disable-health-checks) these health checks.
* *Added:* Audit Logs now support [filtering events](/audit-logs/events/view) by more fields, so you can find events that occurred in a specific region, are associated with operations that have a specific status, etc.
* *Updated:* The navigation in the [web console](https://console.nebius.com) has undergone a major overhaul to group and present services and resources more clearly.
* *Updated:* The upload and download speed for the Standard [storage class](/object-storage/storage-classes) in Object Storage is now limited to 20 GBps per tenant in all regions.