AWS offers access to foundation models from Anthropic, Meta, Mistral, and Amazon through Amazon Bedrock, helping you customize generative AI solutions for chatbots, code assistants, and more. Use a simple 8-step decision framework to navigate the customization spectrum, from using existing models to training custom ones, to optimize performance and cost-effectiveness.
Atlas Building Composites, a spinout of MIT, uses AI-powered robotic manufacturing to turn single-use plastics into durable building materials. The company aims to build 1 billion homes while addressing plastic pollution by recycling bottles into buildings.
Perplexity introduces Portable Computer for Windows PCs, powered by NVIDIA GPUs, allowing local handling of sensitive data and multistep tasks. Users can seamlessly integrate local and cloud AI, simplifying complex tasks and enhancing productivity across various industries.
Summary: The AWS Machine Learning Blog details a solution for automating replenishment in retail, using Databricks and Amazon Quick to predict demand, detect surges, and place orders efficiently. The innovative loop system seamlessly connects forecasts with supplier availability, streamlining the ordering process for retailers.
AI-GUIDE, developed by MIT Lincoln Laboratory and MGH, wins FLC award. Portable device improves medical outcomes for military medics and civilians, with FDA Breakthrough Device Designation.
Implementing Gradient Boost regression with Blind Trees learners showed no significant improvement over standard approaches in predicting numeric values, despite the potential speed and regularization benefits of Blind Trees. The experiment highlights the complexity of explaining machine learning problems and the importance of understanding different regression techniques in the field.
TwelveLabs Marengo Embed 3.0 now available in Amazon Bedrock Knowledge Bases for natural language search over video, audio, and image content. Marengo Embed 3.0 offers a fully managed multimodal embedding model for seamless semantic search experiences.
Agent Evaluation Metric (AEM) reveals root cause of failure in multi-turn conversations. Existing metrics fail to track quality breakdown turn by turn.
Amazon SageMaker Inference introduces prefix-aware routing to optimize large language model (LLM) processing. It reduces time-to-first-token (TTFT) by up to 77% and increases throughput by up to 16%, improving KV cache hit rates from 25% to over 80%.
GeForce NOW introduces WARDOGS, Valheim 1.0 Deep North update, and Bus Simulator 27, offering top PC games without hardware limitations. Enjoy high-performance gaming across devices with no downloads required.
AvioBook, a Thales Group Company, helps airlines save money by optimizing turnaround times with their software AvioBook Connect. Their Connected Analytics feature uses Amazon Bedrock AgentCore to provide real-time answers to operational questions, improving efficiency and reducing delays.
Automating custom permissions in Amazon Quick ensures fine-grained access control for users. Four architectural patterns automate custom permissions at key stages of the user lifecycle.
TorchServe is no longer maintained, leaving teams to handle security patches and compatibility issues. AWS introduces Ray Serve DLC for efficient model inference deployment on AWS platforms, streamlining the process with pre-built Docker images.
NVIDIA AI for Media enhances broadcast and production workflows with real-time intelligence at IBC 2026. Dalet, TwelveLabs, and Wowza integrate NVIDIA's Synthetic Video Detector for content verification and compliance in media and entertainment industries.
Alibaba's Qwen team released Qwen3.8-2.4T-A95B, the first open-weight Qwen-Max-class model with 2.4 trillion total parameters for demanding agentic and reasoning tasks. Deploying on Amazon SageMaker HyperPod enables customized inference behavior and native Multi-Token Prediction for speculative decoding.