live wire
▸AI · vLLM Recipes maps Muse Spark Open to single-node Blackwell and AMD deployment (Oct. 16)vLLM▸PLATFORM · Clang 23 bytecode interpreter cuts allocations and expands constant-expression supportRed Hat Developer▸PLATFORM · RHEL becomes first-party on OCI; OpenShift subscriptions reach Oracle MarketplaceRed Hat▸AI · vLLM-Omni merges native MOSS-Image serving with parallelism and offloadvLLM▸AI · vLLM adds System One endpoint for Decision 3.0 multimodal routing models (Oct. 15)vLLM▸SECURITY · Red Hat rates OpenShift AI 3.4.5 update Critical; advisory lists hundreds of CVEsRed Hat Product Security▸SECURITY · Red Hat’s first Security Signals report shifts focus from CVE counts to risk paths (Oct. 15)Red Hat▸PLATFORM · IBM adds non-destructive FlexVolume-to-CSI migration tool for OpenShift object storageIBM Cloud▸vLLM tests agent-built Kimi K3 decode path at up to 1.9× speedup; PR remains openvLLM▸AI · LeaderWorkerSet 1.0 changes rollout defaults for disaggregated AI workloadsKubernetes SIGs▸SECURITY · Red Hat flags active exploitation of kernel flaw affecting OpenShift CoreOS; no fix listedRed Hat Product Security▸AI · Red Hat sets Dec. 9 showcase for AI 3.6 agent governance and model servicesRed Hat▸vLLM 0.32.0 drops Python 3.10, restricts CORS defaults and adds security fixesvLLM▸SECURITY · Red Hat tracks Moderate Hono auth bypass across OpenShift AI, Dev Spaces and AnsibleRed Hat Product Security▸AI · vLLM Recipes maps Muse Spark Open to single-node Blackwell and AMD deployment (Oct. 16)vLLM▸PLATFORM · Clang 23 bytecode interpreter cuts allocations and expands constant-expression supportRed Hat Developer▸PLATFORM · RHEL becomes first-party on OCI; OpenShift subscriptions reach Oracle MarketplaceRed Hat▸AI · vLLM-Omni merges native MOSS-Image serving with parallelism and offloadvLLM▸AI · vLLM adds System One endpoint for Decision 3.0 multimodal routing models (Oct. 15)vLLM▸SECURITY · Red Hat rates OpenShift AI 3.4.5 update Critical; advisory lists hundreds of CVEsRed Hat Product Security▸SECURITY · Red Hat’s first Security Signals report shifts focus from CVE counts to risk paths (Oct. 15)Red Hat▸PLATFORM · IBM adds non-destructive FlexVolume-to-CSI migration tool for OpenShift object storageIBM Cloud▸vLLM tests agent-built Kimi K3 decode path at up to 1.9× speedup; PR remains openvLLM▸AI · LeaderWorkerSet 1.0 changes rollout defaults for disaggregated AI workloadsKubernetes SIGs▸SECURITY · Red Hat flags active exploitation of kernel flaw affecting OpenShift CoreOS; no fix listedRed Hat Product Security▸AI · Red Hat sets Dec. 9 showcase for AI 3.6 agent governance and model servicesRed Hat▸vLLM 0.32.0 drops Python 3.10, restricts CORS defaults and adds security fixesvLLM▸SECURITY · Red Hat tracks Moderate Hono auth bypass across OpenShift AI, Dev Spaces and AnsibleRed Hat Product Security
upstreambeat.ai
releaseAI

RHEL AI 3.5 moves the appliance forward—and leaves 1.5 upgrades behind

The GA refresh packages vLLM 0.24 and new accelerator stacks, but operators on RHEL AI 1.5 cannot upgrade directly.

RHEL AI 3.5 upgrade path splits from 1.5.
Side by side: what changed
By The Release Desk· Sep 22, 2026the quick take — two AI hosts go live when you do

Red Hat released Red Hat Enterprise Linux AI 3.5 on September 22 through product-enhancement advisories, while the product documentation identifies the generally available image version as 3.5.0. The release moves the appliance to newer inference, optimization and accelerator components, but it also creates a hard planning boundary for operators still on the 1.5 line. (Red Hat advisory; release notes)

What changed

RHEL AI 3.5.0 is published as bootc images for general CUDA and ROCm deployments and for AWS, Azure and Google Cloud variants; Red Hat also lists a CUDA disk image as GA. The Gaudi bootc image remains Technology Preview. (release notes)

The release packages Red Hat AI Inference 3.5.0 with upstream vLLM 0.24.0. It also packages Red Hat AI Model Optimization Toolkit 3.5.0 with upstream LLM Compressor 0.12.0. (release notes)

The accelerator stack advances to AMD ROCm 7.14 using the TheRock framework, NVIDIA CUDA 13.0.2 and Intel Gaudi software 1.24.1-482. Red Hat specifies NVIDIA driver 580.173.02 as the minimum for CUDA 13.0.2; Gaudi support remains Technology Preview. (release notes)

Who it affects

Operators on RHEL AI 3.0 or later can move through the supported 3.x path to 3.5. Red Hat explicitly says there is no direct upgrade from RHEL AI 1.5 to 3.5, so 1.5 estates need a migration or replacement plan rather than treating this as an ordinary in-place update. (release notes)

The accelerator choice also changes the included toolset. Red Hat says the ROCm bootc image does not contain the Model Optimization Toolkit because that toolkit is not supported on AMD ROCm accelerators. (release notes)

What to do

Before deploying, select the image that matches the target cloud and accelerator, verify that NVIDIA systems meet the minimum driver level, and treat the Gaudi image as non-production Technology Preview software. Operators on 1.5 should plan a redeployment or staged migration; operators on 3.0 and later should follow the documented intermediate upgrade path. (release notes)

sources

Filed by The Release Desk. Corrections: desk@upstreambeat.ai · Our standards →

comments · 0

    Comments are moderated before they appear. Your email is used once to confirm it is you — never shown, never sold. Corrections and questions get an answer from the desk when we have one.

    more from the desk