Releasesby IbisValidate GPU cluster health with Cluster Health Checks
Update2 days agoUnverifiedAdded Oct 24, 2026
Oracle Cloud Infrastructure introduces Cluster Health Checks (CHC) to help you identify unhealthy GPU nodes and validate cluster readiness before running AI and high-performance computing workloads. CHC provides two kinds of health checks: Passive checks (through Oracle Cloud Agent) evaluate GPU, driver, and network health on supported instances.
Topics: AI agents, Governance, Performance
More Oracle Cloud Infrastructure releases
Every Oracle Cloud Infrastructure releaseAdd file storage to Generative AI applications
Hosted applications in OCI Generative AI now support persistent file storage. Store model artifacts, intermediate results, checkpoints, and application state at a configured container path across restarts and redeployments. Use the Console or API to add up to 50 GB of file storage when creating a hosted application.
Create Generative AI applications for public content
You can now select No authentication when creating a hosted application in OCI Generative AI. Use this option to serve public content or run applications that manage their own authentication. Anyone with access to the application endpoint can send requests without platform-managed authentication.
Device Data FHIR resource management and Subscription status enhancements
This Oracle Cloud Infrastructure (OCI) Device Data FHIR Service release includes these capabilities: Create and restore resources with PUT: Use client-assigned logical IDs to create resources, update existing resources, and restore logically deleted resources in FHIR R4 and R6 Ballot 4.
OCI Kubernetes Engine (OKE) support for GPU maintenance event repair
You can now configure OKE to coordinate maintenance event repair for eligible bare metal GPU worker nodes in enhanced clusters. When the GPU Health Monitoring add-on reports an eligible critical HPC fault, OKE correlates the fault with an OCI Compute maintenance event.
Use Sandboxes in OCI Generative AI
Sandboxes are now available in OCI Generative AI. A sandbox provides a managed, temporary Linux workspace where developers and AI agents can run code and shell commands. Each sandbox has its own file system, network namespace, and process space. Use the OCI CLI or SDK to: Create sandboxes within a Generative AI project.
OCI Generative AI adds CPU-only access in 30 commercial regions
OCI Generative AI now supports CPU-only access to on-demand models through routing profiles in 30 OC1 commercial regions. Call supported models hosted in other regions from your region’s playground or through the API.
Import Alibaba Qwen3.6-35B-A3B-FP8 into OCI Generative AI
You can now import the Qwen/Qwen3.6-35B-A3B-FP8 model into OCI Generative AI, create endpoints for it, and use it for image and text understanding. In OCI Generative AI, use the following details for this model: Model capability: IMAGE_TEXT_TO_TEXT Minimum dedicated AI cluster unit shapes: A100_40G_X2 , A100_80G_X1 , H100_X1 , and H200_X1 For the complete...
Import NVIDIA Nemotron 3.5 Super VL NVFP4 into OCI Generative AI
You can now import the nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4 model into OCI Generative AI, create endpoints for it, and use it for image and text understanding.
Also shipped on Oct 22, 2026
Oracle in October 2026Trace and evaluate Cursor agents with Langfuse, powered by ClickHouse
Users can now connect Cursor to Langfuse to trace agent turns and analyze model and tool calls, with ClickHouse providing the underlying analytics infrastructure.
The AI.TRANSLATE function lets you translate text strings between languages
The AI.TRANSLATE function lets you translate text strings between languages. AI.TRANSLATE supports standard Neural Machine Translation (NMT) workloads and Translation LLM (TLLM) capabilities. This function is generally available (GA).
Replit Monitoring Agent: Your very own software engineering team
Replit Monitoring Agent is an autonomous tool that investigates production alerts, identifies causes, applies fixes, and deploys them to production for internal apps.
Microsoft Entra ID connector is available in Beta
The managed Microsoft Entra ID connector is now available in Beta in Databricks Lakeflow Connect. Use the connector to ingest Microsoft Entra ID users, groups, applications, service principals, directory roles, identity configuration, and audit and sign-in logs into Databricks.
You can configure Direct VPC ingress for Cloud Run instances ( Preview )
You can configure Direct VPC ingress for Cloud Run instances ( Preview ). Direct VPC ingress lets Cloud Run instances on a VPC network use their own ephemeral private IP address to communicate with each other and with other resources on the VPC network.
Gemini Enterprise: Single-user quota limits
Gemini Enterprise administrators can configure single-user quota limits on the Usage & Spending > Usage page in the Google Cloud console to prevent an individual user from using all of an edition's shared subscription quota pool.