AWS Machine Learning Blog

Category: Advanced (300)

Talk to your slide deck using multimodal foundation models on HAQM Bedrock – Part 3

In Parts 1 and 2 of this series, we explored ways to use the power of multimodal FMs such as HAQM Titan Multimodal Embeddings, HAQM Titan Text Embeddings, and Anthropic’s Claude 3 Sonnet. In this post, we compared the approaches from an accuracy and pricing perspective.

solution architecture

Accelerating ML experimentation with enhanced security: AWS PrivateLink support for HAQM SageMaker with MLflow

With access to a wide range of generative AI foundation models (FM) and the ability to build and train their own machine learning (ML) models in HAQM SageMaker, users want a seamless and secure way to experiment with and select the models that deliver the most value for their business. In the initial stages of an ML […]

Search enterprise data assets using LLMs backed by knowledge graphs

In this post, we present a generative AI-powered semantic search solution that empowers business users to quickly and accurately find relevant data assets across various enterprise data sources. In this solution, we integrate large language models (LLMs) hosted on HAQM Bedrock backed by a knowledge base that is derived from a knowledge graph built on HAQM Neptune to create a powerful search paradigm that enables natural language-based questions to integrate search across documents stored in HAQM Simple Storage Service (HAQM S3), data lake tables hosted on the AWS Glue Data Catalog, and enterprise assets in HAQM DataZone.

Embodied AI Chess with HAQM Bedrock

In this post, we demonstrate Embodied AI Chess with HAQM Bedrock, bringing a new dimension to traditional chess through generative AI capabilities. Our setup features a smart chess board that can detect moves in real time, paired with two robotic arms executing those moves. Each arm is controlled by different FMs—base or custom. This physical implementation allows you to observe and experiment with how different generative AI models approach complex gaming strategies in real-world chess matches.

Efficiently train models with large sequence lengths using HAQM SageMaker model parallel

In this post, we demonstrate how the HAQM SageMaker model parallel library (SMP) addresses this need through support for new features such as 8-bit floating point (FP8) mixed-precision training for accelerated training performance and context parallelism for processing large input sequence lengths, expanding the list of its existing features.

How Crexi achieved ML models deployment on AWS at scale and boosted efficiency

Commercial Real Estate Exchange, Inc. (Crexi), is a digital marketplace and platform designed to streamline commercial real estate transactions. In this post, we will review how Crexi achieved its business needs and developed a versatile and powerful framework for AI/ML pipeline creation and deployment. This customizable and scalable solution allows its ML models to be efficiently deployed and managed to meet diverse project requirements.

How 123RF saved over 90% of their translation costs by switching to HAQM Bedrock

This post explores how 123RF used HAQM Bedrock, Anthropic’s Claude 3 Haiku, and a vector store to efficiently translate content metadata, significantly reduce costs, and improve their global content discovery capabilities.

Connect SharePoint Online to HAQM Q Business using OAuth 2.0 ROPC flow authentication

In this post, we explore how to integrate HAQM Q Business with SharePoint Online using the OAuth 2.0 ROPC flow authentication method. We provide both manual and automated approaches using PowerShell scripts for configuring the required Azure AD settings. Additionally, we demonstrate how to enter those details along with your SharePoint authentication credentials into the HAQM Q console to finalize the secure connection.

Accelerating Mixtral MoE fine-tuning on HAQM SageMaker with QLoRA

In this post, we demonstrate how you can address the challenges of model customization being complex, time-consuming, and often expensive by using fully managed environment with HAQM SageMaker Training jobs to fine-tune the Mixtral 8x7B model using PyTorch Fully Sharded Data Parallel (FSDP) and Quantized Low Rank Adaptation (QLoRA).