AWS Moves Blackwell GPU Compute to the Edge in Los Angeles

Amazon Web Services

Apr 24, 2026 · Updated May 2, 2026

AWS announced the general availability of Amazon EC2 G7e instances in its Los Angeles Local Zone. By placing NVIDIA Blackwell-class GPUs physically closer to end users, the update enables sub-millisecond latency for real-time VFX rendering and agentic AI loops.

Amazon Web Services (AWS) launched the general availability of Amazon EC2 G7e instances in its Los Angeles Local Zone. These instances feature NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs and 5th generation Intel Xeon Scalable processors. This expansion brings high-performance GPU compute to the network edge.

GPU: NVIDIA RTX PRO 6000 Blackwell Server Edition
CPU: 5th Gen Intel Xeon Scalable
Local Zone: Los Angeles (us-west-2-lax-1b)
Availability: General Availability
Purchase Options: On Demand, Savings Plans

This move follows a broader industry shift toward distributed AI grids that reduce the latency bottleneck of centralized data centers. For agentic AI systems requiring rapid multi-step reasoning, physical proximity is critical. It mirrors the pattern of moving inference to the edge to support real-time autonomous operations.

You can now deploy Large Language Models (LLMs) (AI trained to understand and generate text) and post-production pipelines in the us-west-2-lax-1b zone. This setup supports real-time 3D rendering and VFX composition with low-latency local storage. To start, opt-in via the AWS Global View console; instances are available through On Demand and Savings Plans.

View the full update on aws.amazon.com

AWS AI

@AWSAIApr 21

Amazon EC2 G7e instances are now generally available in AWS Local Zones in Los Angeles. High-performance GPU compute moves closer to end users with instances accelerated by @nvidia RTX PRO 6000 Blackwell Server Edition GPUs. With Amazon EC2 G7e instances, run studio workstation workloads and post-production pipelines, or deploy LLMs and agentic AI at the edge to meet low latency requirements. https://t.co/1H2bu6WrpX

View on X

Still wondering? A few quick answers below.

Amazon EC2 G7e instances are high-performance virtual machines designed for graphics-intensive and AI workloads. They feature NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs and 5th generation Intel Xeon Scalable processors. These instances are optimized for tasks requiring a balance of powerful GPU compute and low-latency access, such as real-time rendering and deploying large language models.

As of April 2026, Amazon EC2 G7e instances are generally available in AWS Local Zones in Los Angeles, California. Specifically, users can access them in the us-west-2-lax-1b zone. This geographic expansion allows businesses in the Southern California region to run demanding GPU workloads physically closer to their end users to minimize network latency.

These instances are accelerated by NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, which provide advanced graphics and AI inference capabilities. They also utilize 5th generation Intel Xeon Scalable processors, also known as Emerald Rapids. This combination of hardware supports enhanced real-time rendering, 2D and 3D visual effects composition, and the execution of complex agentic AI loops.

To use G7e instances in Los Angeles, you must first opt-in to the us-west-2-lax-1b Local Zone through the AWS Global View console. Once enabled, you can launch and manage the instances using the Amazon EC2 console, AWS Command Line Interface, or AWS SDKs. They are available through standard On Demand pricing or via Savings Plans.

G7e instances are built for creative and AI workloads that require low latency. Creative uses include visual effects editorial, color correction, and post-production finishing. For AI, they are used to deploy large language models and agentic AI systems at the edge, where immediate response times are necessary for interactive applications and real-time data processing.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

See all AI news & updates from AWS →

Keep reading

AWS First Major Cloud Provider Supporting NVIDIA RTX PRO 4500 Blackwell GPUs

Amazon EC2 will add instances accelerated by NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs, making AWS the first major cloud provider to announce support. These instances suit workloads like conversational AI, data analytics, content generation, and graphics.

NVIDIA Dynamo 1.0 Ships as Open-Source Inference OS for AI Factories

NVIDIAMar 18

NVIDIA Dynamo 1.0 Ships as Open-Source Inference OS for AI Factories

NVIDIA Dynamo 1.0 reaches production as open-source software that orchestrates GPU clusters for AI inference at data center scale. It boosts Blackwell GPU inference performance by up to 7x and already runs on AWS, Azure, Google Cloud, and OCI.

Qwen Sets 580 TPS Record for Agentic Workloads on Blackwell GPUs

QwenMay 27

Qwen Sets 580 TPS Record for Agentic Workloads on Blackwell GPUs

Qwen achieved a record 580 tokens per second running its Qwen3.5-397B-A17B model on NVIDIA Blackwell GPUs using the TokenSpeed inference engine. The optimization targets agentic workloads, where multi-turn reasoning and tool-calling typically suffer from high latency. By combining a hybrid attention architecture with deep kernel fusion, the system maintains high throughput even as context scales to one million tokens.

Perplexity Benchmarks Blackwell Performance for High Throughput Large Model Inference

PerplexityMay 12

Perplexity Benchmarks Blackwell Performance for High Throughput Large Model Inference

Perplexity published research showing that NVIDIA's GB200 Blackwell architecture nearly halves communication latency for large Mixture-of-Experts models compared to the previous generation. The findings suggest that Blackwell is a primary platform for reducing the cost and latency of serving frontier-scale AI search.

What are Amazon EC2 G7e instances?

Where are G7e instances available?

What hardware powers the AWS G7e instances?

How do I access G7e instances in the Los Angeles Local Zone?

What are the primary use cases for G7e instances at the edge?

Keep reading

AWS First Major Cloud Provider Supporting NVIDIA RTX PRO 4500 Blackwell GPUs

AWS First Major Cloud Provider Supporting NVIDIA RTX PRO 4500 Blackwell GPUs

NVIDIA Dynamo 1.0 Ships as Open-Source Inference OS for AI Factories

NVIDIA Dynamo 1.0 Ships as Open-Source Inference OS for AI Factories

Qwen Sets 580 TPS Record for Agentic Workloads on Blackwell GPUs

Qwen Sets 580 TPS Record for Agentic Workloads on Blackwell GPUs

Perplexity Benchmarks Blackwell Performance for High Throughput Large Model Inference

Perplexity Benchmarks Blackwell Performance for High Throughput Large Model Inference

Keep reading

AWS First Major Cloud Provider Supporting NVIDIA RTX PRO 4500 Blackwell GPUs

AWS First Major Cloud Provider Supporting NVIDIA RTX PRO 4500 Blackwell GPUs

NVIDIA Dynamo 1.0 Ships as Open-Source Inference OS for AI Factories

NVIDIA Dynamo 1.0 Ships as Open-Source Inference OS for AI Factories

Qwen Sets 580 TPS Record for Agentic Workloads on Blackwell GPUs

Qwen Sets 580 TPS Record for Agentic Workloads on Blackwell GPUs

Perplexity Benchmarks Blackwell Performance for High Throughput Large Model Inference

Perplexity Benchmarks Blackwell Performance for High Throughput Large Model Inference