All news with #product update tag

🚀 The NVIDIA Run:ai Model Streamer now supports native Google Cloud Storage access, accelerating model load and inference startup for vLLM workloads on GKE. By streaming tensors directly from Cloud Storage into GPU memory and using distributed, NVLink-aware transfers, the streamer dramatically reduces cold-start latency and idle GPU time. Enabling it in vLLM is a single-flag change and it can leverage GKE Workload Identity for secure, keyless access.

⚠️ Microsoft is investigating a known issue that prevents customers from downloading Microsoft 365 desktop apps from the Microsoft 365 homepage, with failures reported since November 2. The company says a recent service update introduced a code defect affecting the license check process, and it has tagged the situation as an incident. A fix has been developed and is being validated in Microsoft's internal environment, and the company promised an update on deployment timing by 6:30 PM UTC. Microsoft is also addressing a separate issue causing some users to be unable to open Excel attachments in the new Outlook client due to filename encoding errors.

🚀 Amazon Bedrock now exposes an OpenAI-compatible Responses API on new service endpoints, enabling asynchronous inference for long-running workloads, streaming and non-streaming modes, and automatic stateful conversation reconstruction so developers no longer must resend full histories. The endpoints provide Chat Completions with reasoning-effort support for models served by Mantle, Amazon’s distributed inference engine. Integration requires only a base URL change for OpenAI SDK–compatible code, and support starts today for OpenAI’s GPT OSS 20B and 120B models, with additional models coming soon.

🚀 Amazon SageMaker HyperPod now supports checkpointless training, a foundational capability that eliminates the need for checkpoint-based, job-level restarts for distributed model training. Checkpointless training preserves forward training state across the cluster, automatically swaps out failed nodes, and uses peer-to-peer state transfer to resume progress, reducing recovery time from hours to minutes. The feature can deliver up to 95% training goodput at very large scale, is available in all Regions where HyperPod runs, and can be enabled with zero code changes for popular recipes or with minimal PyTorch modifications for custom models.

🚀 AWS has announced TypeScript support in preview for the Strands Agents SDK, giving developers a choice between Python and TypeScript for building model-driven AI agents. The TypeScript implementation provides idiomatic, type-safe APIs with async/await and modern JavaScript/TypeScript patterns, and is designed to run in browsers, client applications, and server runtimes such as AWS Lambda and Bedrock AgentCore. AWS also introduced three SDK updates: edge device support is now GA, Strands steering is available experimentally, and Strands evaluations is in preview to help validate agent behavior.

🔧 Amazon Bedrock now supports reinforcement fine-tuning, enabling developers to improve model accuracy without deep ML expertise or large labeled datasets. The service automates the reinforcement fine-tuning workflow and trains models by learning from feedback on multiple candidate responses, improving model judgment about what makes a good reply. AWS reports an average 66% accuracy gain over base models, allowing teams to deploy smaller, faster, and more cost-effective variants while maintaining quality. At launch the feature supports Amazon Nova 2 Lite, and it can be accessed via the Bedrock console or APIs.

⚡ Amazon SageMaker HyperPod now supports elastic training, automatically scaling distributed training jobs to absorb idle accelerators and contract when higher‑priority workloads require resources. This eliminates the manual cycle of halting jobs, reconfiguring parameters, and restarting distributed training, which previously demanded specialized engineering time. Organizations can start training with minimal resources and grow opportunistically, improving cluster utilization and reducing costs. Elastic training can be enabled with zero code changes for public models like Llama and GPT OSS, and requires only lightweight configuration updates for custom architectures.

🔧 Microsoft has published the optional KB5070311 preview cumulative update for Windows 11, delivering 49 non-security fixes and quality improvements. The November preview resolves an explorer.exe and taskbar hang triggered by certain notifications, corrects File Explorer search issues affecting some SMB shares, and addresses an LSASS access-violation instability. Install via Settings → Windows Update or download from the Microsoft Update Catalog; this update advances 25H2 and 24H2 builds to 26200.7309 and 26100.7309 respectively.

🔁 Amazon Bedrock AgentCore Runtime now supports bi-directional streaming, enabling real-time, continuous conversations where agents listen and respond simultaneously and handle interruptions or context shifts mid-turn. This removes stop-start friction in voice and text agents and preserves context across exchanges. Built into AgentCore Runtime, the capability reduces months of engineering work required to implement streaming infrastructure, letting developers focus on agent experiences rather than plumbing. Available in nine AWS Regions with consumption-based pricing.

🧠 Amazon SageMaker AI now offers a serverless MLflow capability that automatically scales to support experiment tracking and model development without infrastructure setup. The service scales up for demanding workloads and scales down during idle periods, reducing operational overhead. Administrators can enable cross-account access via Resource Access Manager (RAM). The feature integrates with SageMaker AI JumpStart, Model Registry, and Pipelines and is offered at no additional charge in select AWS Regions.

🗃️ Amazon S3 Tables now support the Intelligent-Tiering storage class to automatically optimize table storage costs based on access patterns, without impacting performance or adding operational overhead. Data not accessed for 30 days moves to the Infrequent Access tier (≈40% lower cost), and after 90 days moves to Archive Instant Access (≈68% lower cost), enabling up to 80% storage savings. Automated table maintenance (compaction, snapshot expiration, unreferenced file removal) does not trigger tiering, and you can select Intelligent-Tiering per table or set it as the default for new tables in a table bucket across all regions where S3 Tables are available.

🔍 Amazon SageMaker Catalog now exports asset metadata as an Apache Iceberg table via Amazon S3 Tables, enabling teams to query catalog inventory with standard SQL without building custom ETL. The export includes technical fields (resource_id, resource_type), business metadata (asset_name, business_description), ownership details, and timestamps, partitioned by snapshot_date for time travel queries. The dataset appears in SageMaker Unified Studio and is queryable from Amazon Athena, Studio notebooks, AI agents, and BI tools. Available in all supported Regions at no additional SageMaker charge; you pay for S3 Tables storage and Athena queries.

🧭 Amazon CloudWatch now offers AgentCore Evaluations, an automated capability for assessing AI agent quality using real-world interactions. The feature includes 13 pre-built evaluators that measure dimensions such as helpfulness, tool selection, and response accuracy, and it also supports custom model-based scoring. Teams can access unified quality metrics, agent telemetry, and end-to-end traces in CloudWatch dashboards to correlate evaluations with prompts, logs, and traces for faster diagnosis and improvement.

🔧 Microsoft is addressing a bug that prevents some users from opening Excel email attachments in the new Outlook client when filenames contain non‑ASCII characters. The company says the root cause is a missing encoding in the file‑open requests and that a fix has been developed and deployed for validation. While the rollout is still in progress, affected users are advised to use Outlook on the web or download the file to open it locally as a temporary workaround.

🔒 Kaspersky has released a major update to Kaspersky Embedded Systems Security, targeting the unique risks of legacy and resource-constrained devices. The Windows edition introduces a behavioral analysis engine plus Automatic Exploit Prevention, Anti-Cryptor, a Remediation Engine, BadUSB protection, a firewall, and a security status indicator. The Linux edition adds certificate-based allowlisting and Web Threat Protection to simplify safe updates and guard web-enabled embedded devices. Planned Q1 2026 improvements include MDR integration, BadUSB for Linux, and ARM support.

🛡️ usecure has appointed Kevin Lancaster as a Non-Executive Director to accelerate its North American channel expansion. Lancaster, founder of ID Agent and former head of Channel Program, brings deep channel experience and a proven track record of scaling channel-first security and SaaS businesses. He will work with the board and executive team to help usecure become the leading human risk management solution for MSPs, supporting growth across distribution partners and more than 1,800 MSP partners worldwide.

🤖 Amazon Connect has introduced AI-powered predictive insights in public preview, adding five recommendation algorithms to its Customer Profiles. The capabilities analyze interaction history and behavior to power suggestions across self-service and agent interactions, supporting cross-sell, complementary product suggestions, and intelligent product discovery. The preview is available in multiple AWS regions and uses pay-as-you-go pricing for utilized profiles.

🤖 Amazon Connect now offers real-time AI agent assistance and automated contact summarization for Salesforce Contact Center with Amazon Connect (SCC-AC). The capability surfaces customer CRM records and knowledge-base articles to AI agents across voice and chat and provides agents with live transcripts and contextual recommendations. Supervisors gain enhanced call monitoring inside Salesforce, and automated post-contact summaries help agents quickly update cases, reducing administrative burden. Administrators can deploy and configure the integrated solution in minutes; it is available in all AWS Regions where Amazon Connect operates.

🤖 Amazon Connect introduces agentic self-service capabilities that enable AI agents to understand, reason, and take action across voice and messaging channels to automate routine and complex customer-service tasks. Nova Sonic speech models deliver more natural, expressive, and adaptive voice interactions that respond to customer tone, sentiment, and pacing across languages and accents. The feature supports blending deterministic and agentic experiences, automating tasks like order lookup, refunds, and troubleshooting while allowing escalation to live agents and is commercially available in US East (N. Virginia) and US West (Oregon) in English and Spanish, with previews for French, Italian, and German.

🔗 Amazon Connect now lets agents link related contacts — including email replies, call transfers, persistent chats, and queued callbacks — to an existing case using configurable flows. By enabling follow-up contacts to attach automatically, Amazon Connect Cases removes the need for custom logic or manual linking and helps present a unified customer journey to agents. The capability is available in multiple AWS regions; see the Amazon Connect Cases documentation to get started.