All news with #amazon bedrock tag

🚀 Amazon Bedrock Data Automation (BDA) is now generally available in the AWS GovCloud (US-West) Region. BDA automates extraction of actionable insights from unstructured multimodal content—documents, images, video, and audio—helping developers accelerate GenAI-based applications like intelligent document processing and media analysis. It can run standalone or as a parser in Amazon Knowledge Bases RAG workflows and is now offered in eight AWS Regions.

🧮 The Count Tokens API is now available in Amazon Bedrock, enabling users to determine token counts for a prompt or input prior to performing inference. Anthropic’s Claude models are supported at launch and the feature is available in all regions where those models run. This improves cost projection, gives more control over token limits, and reduces the risk of unexpected throttling. It also helps ensure inputs fit within a model's context length for more efficient prompt optimization.

🔓 AWS has made two OpenAI models with open weights — gpt-oss-120b and gpt-oss-20b — automatically available to all Amazon Bedrock users as of August 5, 2025. Users can access them immediately via the Amazon Bedrock console playground or the unified Bedrock API in supported regions. Administrators retain full control and can restrict usage with AWS IAM policies and Service Control Policies.

📹 TwelveLabs Pegasus 1.2 is now available in US East (N. Virginia) and Asia Pacific (Seoul) through Amazon Bedrock. The video-first language model is optimized for long-form content and combines visual, audio, and textual signals to deliver advanced video-to-text generation and temporal understanding. Regional availability reduces latency and simplifies architecture for enterprise video-intelligence applications. To begin, request model access via the Amazon Bedrock console.

🚀 Amazon Bedrock now supports Batch inference for Anthropic Claude Sonnet 4 and OpenAI GPT-OSS (120B, 20B), enabling asynchronous processing of large workloads at approximately 50% of on-demand inference cost. The update targets bulk scenarios such as document analysis, large-scale summarization, content generation, and structured data extraction, and is optimized to deliver higher overall batch throughput on these newer models. Batch progress and workload metrics — including pending and processed records, tokens per minute, and Claude-specific pending tokens — are exposed at the AWS account level via Amazon CloudWatch.

🧠 Amazon Neptune now integrates with Cognee to provide graph-native memory for agentic generative AI applications. The integration enables developers to use Amazon Neptune Analytics as the persistent graph and vector store behind Cognee’s memory layer, supporting large-scale memory graphs, long-term memory, and multi-hop reasoning. Hybrid retrieval across graph, vector, and keyword modalities helps agents deliver more personalized, cost-efficient, and context-aware experiences; documentation and a sample notebook are available to accelerate adoption.