AI news for builders and product teamsUpdated Nov 17, 2026, 14:00 UTC
AWS Machine Learning Blog
First-party releases and research from AWS Machine Learning Blog. Headlines and excerpts link to the original articles.
Latest stories
Newest first
uniopen customized Amazon Nova 2 Lite for its retail content moderation policies using supervised fine-tuning with LoRA in Amazon SageMaker AI plus a prompt-level output change. Per Behavior Macro F1 rose from 0.5852 to 0.8550 and Subject Type Macro F1 from 0.4162 to 0.8491, both above production targets.

Condé Nast partnered with the AWS Generative AI Innovation Center to build multimodal video discovery on Amazon Bedrock and Amazon OpenSearch Service, using the TwelveLabs Marengo embedding model. Condé Nast reported a 99.2 percent reduction in content discovery time, from 250 minutes to about 2 minutes per task.

xAI's Grok 4.7 is now available on Amazon Bedrock, offering a 500K token context window and four configurable reasoning effort levels: low, medium, high, and xhigh. It is served on the bedrock-runtime endpoint through cross-Region inference profiles and supports the Responses, Chat Completions, and Converse APIs.

AWS announced Claude Sonnet 5.5 is available on Amazon Bedrock and Claude Platform on AWS, described as a more efficient Sonnet model for focused coding and knowledge work with lower cost per task for most work at faster speed.

AWS published a tutorial on deploying the Qwen3-TTS text-to-speech model on Amazon SageMaker AI using the vLLM-Omni Deep Learning Container, streaming text in and 24 kHz PCM audio out over one bidirectional WebSocket connection through a Gradio app.

An AWS Machine Learning Blog post describes deploying two Amazon SageMaker AI endpoints from the same AWS vLLM-Omni Deep Learning Container: a real-time endpoint for FLUX.2-klein-4B image generation and an asynchronous endpoint for Wan2.1-VACE-1.3B video generation. The sample turns a text prompt into an image, then animates it into a short MP4 stored in Amazon S3.

AWS published a technical post describing an agent-driven synthetic monitoring solution built on Amazon Nova Act and Amazon Bedrock AgentCore, with a sample repository. The post covers architecture, deployment, cost and security considerations, and reports over 90% accuracy for Nova Act in early enterprise customer browser workflow use cases.

An AWS Machine Learning Blog post describes patterns for managing Amazon Textract adapter lifecycles across AWS accounts, including infrastructure templates, CI/CD promotion, security configuration, and a pre-classification routing approach. It details cross-account copy via AWS Support or a centralized hub account model, and externalizing adapter IDs in Parameter Store.

AWS published a post describing an architecture that combines Amazon EKS, Elastic Fabric Adapter (EFA), and DeepEP to support large-scale MoE reinforcement learning. Across 48 P5en instances running a super-sparse MoE model, enabling DeepEP over EFA increased aggregate RL rollout throughput by 40 percent.

An AWS Machine Learning Blog post walks through running SkyRL, an open-source reinforcement learning framework, on Amazon SageMaker HyperPod to train a Qwen3-VL-8B vision-language model to navigate visual mazes with GRPO. Starting from the VisGym SFT checkpoint, the post reports the maze solve rate rose from 43.75% to more than 95% on a fixed 64-maze evaluation set.

AWS published a post detailing five quality assurance techniques in its NarrateAI assistant built on Amazon Bedrock, used by over 4,000 AWS executive leaders. The techniques are adaptive pipeline orchestration, cross-account multi-model failover, real-time streaming evaluation, composite evaluation framework and data accuracy verification, which AWS says achieve approximately 99 percent numerical accuracy.

AWS published a walkthrough for deploying the public Qwen3-TTS-12Hz-1.7B-Base text-to-speech model from Amazon SageMaker JumpStart to a fully managed real-time inference endpoint, including voice cloning from a short reference clip and transcript.

Datacor integrated Amazon Quick Sight into its TrackAbout solution to give gas and welding distributors self-service rental analytics with embedded dashboards and natural language querying. The solution, released in November 2025, is used by more than 50 customer organizations and over 60 business users.