Announcements

Introducing Amazon Polly Bidirectional Streaming: Real-time speech synthesis for conversational AI

Introducing Amazon Polly Bidirectional Streaming: Real-time speech synthesis for conversational AI

Introducing Amazon Polly Bidirectional Streaming: Real-time speech synthesis for conversational AI

Deven Goratela 30 March 2026

Building natural conversational experiences requires speech synthesis that keeps pace with real-time interactions. Today, we’re excited to...

Run Generative AI inference with Amazon Bedrock in Asia Pacific (New Zealand)

Run Generative AI inference with Amazon Bedrock in Asia Pacific (New Zealand)

Run Generative AI inference with Amazon Bedrock in Asia Pacific (New Zealand)

Deven Goratela 28 March 2026

Kia ora! Customers in New Zealand have been asking for access to foundation models (FMs) on Amazon...

Run NVIDIA Nemotron 3 Super on Amazon Bedrock

Run NVIDIA Nemotron 3 Super on Amazon Bedrock

Run NVIDIA Nemotron 3 Super on Amazon Bedrock

Deven Goratela 20 March 2026

Nemotron 3 Super is now available as a fully managed and serverless model on Amazon Bedrock, joining...

Our First 2026 Heroes Cohort Is Here!

Our First 2026 Heroes Cohort Is Here!

Our First 2026 Heroes Cohort Is Here!

Deven Goratela 18 March 2026

We’re thrilled to celebrate three exceptional developer community leaders as AWS Heroes. These individuals represent the heart...

AWS Weekly Roundup: Amazon S3 turns 20, Amazon Route 53 Global Resolver general availability, and more (March 16, 2026)

AWS Weekly Roundup: Amazon S3 turns 20, Amazon Route 53 Global Resolver general availability, and more (March 16, 2026)

AWS Weekly Roundup: Amazon S3 turns 20, Amazon Route 53 Global Resolver general availability, and more (March 16, 2026)

Deven Goratela 16 March 2026

Twenty years ago this past week, Amazon S3 launched publicly on March 14, 2006. While Amazon Simple...

P-EAGLE: Faster LLM inference with Parallel Speculative Decoding in vLLM

P-EAGLE: Faster LLM inference with Parallel Speculative Decoding in vLLM

P-EAGLE: Faster LLM inference with Parallel Speculative Decoding in vLLM

Deven Goratela 15 March 2026

EAGLE is the state-of-the-art method for speculative decoding in large language model (LLM) inference, but its autoregressive...

Improve operational visibility for inference workloads on Amazon Bedrock with new CloudWatch metrics for TTFT and Estimated Quota Consumption

Improve operational visibility for inference workloads on Amazon Bedrock with new CloudWatch metrics for TTFT and Estimated Quota Consumption

Improve operational visibility for inference workloads on Amazon Bedrock with new CloudWatch metrics for TTFT and Estimated Quota Consumption

Deven Goratela 14 March 2026

As organizations scale their generative AI workloads on Amazon Bedrock, operational visibility into inference performance and resource...

Efficiently serve dozens of fine-tuned models with vLLM on Amazon SageMaker AI and Amazon Bedrock

Efficiently serve dozens of fine-tuned models with vLLM on Amazon SageMaker AI and Amazon Bedrock

Efficiently serve dozens of fine-tuned models with vLLM on Amazon SageMaker AI and Amazon Bedrock

Deven Goratela 2 March 2026

Organizations and individuals running multiple custom AI models, especially recent Mixture of Experts (MoE) model families, can...

Customize AI agent browsing with proxies, profiles, and extensions in Amazon Bedrock AgentCore Browser

Customize AI agent browsing with proxies, profiles, and extensions in Amazon Bedrock AgentCore Browser

Customize AI agent browsing with proxies, profiles, and extensions in Amazon Bedrock AgentCore Browser

Deven Goratela 14 February 2026

AI agents that browse the web need more than basic page navigation. Our customers tell us they...

NVIDIA Nemotron 3 Nano 30B MoE model is now available in Amazon SageMaker JumpStart

NVIDIA Nemotron 3 Nano 30B MoE model is now available in Amazon SageMaker JumpStart

NVIDIA Nemotron 3 Nano 30B MoE model is now available in Amazon SageMaker JumpStart

Deven Goratela 12 February 2026

Today we’re excited to announce that the NVIDIA Nemotron 3 Nano 30B model with 3B active parameters...