Sold by: Meetrix.io
Deployed on AWS
vLLM is a high-throughput inference and serving engine for large language models. This preconfigured Amazon Machine Image deploys a production-ready vLLM server with an OpenAI-compatible API on a GPU EC2 instance, so you can serve open-source models such as Llama, Qwen, Mistral, and Gemma from your own infrastructure in minutes. It ships with NVIDIA drivers, the NVIDIA Container Toolkit, an Nginx reverse proxy, and automatic Let's Encrypt SSL, and uses PagedAttention and continuous batching to maximize tokens per second and GPU utilization. Point any OpenAI SDK at your endpoint and start generating.
Overview
vLLM is an open-source library for high-throughput large language model inference and serving, built around PagedAttention and continuous batching. This preconfigured Amazon Machine Image runs the vLLM OpenAI-compatible API server on a GPU EC2 instance, giving developers, ML teams, and SaaS platforms a fast, private, and cost-controlled way to serve open-weight models without managing a complex inference stack.
vLLM Step by step Installation Guide: https://meetrix.io/blogs/vllm-developer-guide/
For custom projects of your choice with personalized support, contact us at https://meetrix.io/contact-us
High-Throughput Serving: PagedAttention and continuous batching keep the GPU saturated, delivering high tokens-per-second and efficient memory use even under concurrent load.
OpenAI-Compatible API: A drop-in /v1 REST interface for chat, completions, and embeddings works with existing OpenAI SDKs and tooling by simply changing the base URL.
Bring Your Own Model: Serve any compatible model from Hugging Face, including Llama, Qwen, Mistral, Gemma, and Phi, and switch models with a single configuration change.
GPU-Ready Out of the Box: NVIDIA drivers and the NVIDIA Container Toolkit are preinstalled and validated, with support for G4, G5, and G6 GPU instance families.
Secure by Default: An Nginx reverse proxy with automatic Let's Encrypt SSL via the Route 53 DNS challenge provides HTTPS access, while SSH is locked to your chosen IP range.
Self-Hosted and Private: Prompts, responses, and model weights stay inside your own account and VPC, giving you full control over data residency and cost.
Fast First Boot: The vLLM container image is pre-pulled into the AMI, so the stack completes in minutes and only model weights download on first start.
Version: v0.28.0
Operating System: Ubuntu 26.04
Deploy a production-grade LLM inference endpoint on your own GPU infrastructure with predictable pricing and expert support from Meetrix. Full setup instructions are in the developer guide: https://meetrix.io/blogs/vllm-developer-guide/
Highlights
- High-Throughput Inference: PagedAttention and continuous batching keep your GPU saturated, delivering high tokens-per-second and efficient memory use even under heavy concurrent request load.
- Standards-Compatible API: A drop-in /v1 REST endpoint for chat, completions, and embeddings works with popular LLM SDKs and tooling by changing only the base URL, so integration takes minutes.
- Private GPU Deployment: Model weights, prompts, and responses stay inside your own account and VPC on preconfigured NVIDIA G4, G5, and G6 instances, with automatic HTTPS and IP-restricted SSH access.
Details
Sold by
Delivery method
Delivery option
vLLM Deployment with GPU-Accelerated Inference
Latest version
Operating system
Ubuntu 26.04
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Cost/hour |
|---|---|
g4dn.xlarge | $0.019 |
g4dn.2xlarge | $0.024 |
g4dn.4xlarge | $0.039 |
g5.xlarge | $0.019 |
g5.2xlarge | $0.024 |
g5.4xlarge | $0.039 |
g5.12xlarge | $0.048 |
g6.xlarge | $0.019 |
g6.2xlarge | $0.024 |
g6.4xlarge | $0.039 |
Vendor refund policy
We do not currently support refunds, but you can cancel at any time.
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
vLLM Deployment with GPU-Accelerated Inference
This deployment provisions a complete, isolated environment for vLLM: a dedicated VPC with a public subnet, internet gateway, and route table, plus a security group that exposes ports 80, 443, and 8000 while restricting SSH to your chosen IP range. A GPU EC2 instance from the G4, G5, or G6 family is launched from the preconfigured AMI with a 100 GiB gp3 root volume and an attached Elastic IP. On first boot, UserData patches the domain, admin email, Hugging Face model ID, and optional access token into the vLLM configuration, starts the container from the pre-pulled image, and requests a Let's Encrypt certificate through the Route 53 DNS challenge. A scoped IAM role grants the instance permission to update Route 53 records and upload backups to S3. Once the model finishes downloading, the OpenAI-compatible API is available at https:///v1, with interactive documentation at /docs and /redoc.
CloudFormation Template (CFT)
AWS CloudFormation templates are JSON or YAML-formatted text files that simplify provisioning and management on AWS. The templates describe the service or application architecture you want to deploy, and AWS CloudFormation uses those templates to provision and configure the required services (such as Amazon EC2 instances or Amazon RDS DB instances). The deployed application and associated resources are called a "stack."
Version release notes
First Release
Additional details
Usage instructions
Click the "Continue to Subscribe" button. After subscribing, you will need to accept the terms and conditions. Click on "Accept Terms" to proceed. Please wait for a few minutes while the processing takes place. Once it's completed, click on "Continue to Configuration".
IAM Role is set up and configured with the necessary permissions to assume the role for the vLLM service. The IAM Policy is created and responsible for access Route53 and create Letsencrypt SSL certificates and backing up data to the S3 bucket.
Access the application via a browser at http://<your domain name> or http://<Public IPv4 address>. Product will try to create SSL certificate when its deploying, if domain hosted on route53. If the automatic SSL creation unsuccessful then you have to point domain name into server ip, ssh into server and run /root/certificate_generate_standalone.sh. Admin email using for SSL generation.
Resources
Vendor resources
Support
Vendor support
Please contact us through aws@meetrix.io . Please allow up to 12 hours for our support team to address your request.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
REQUIRES PRIVATE OFFER
To purchase LiteLLM Enterprise Self-Hosted, please reach out to sales@berri.ai for a Private Offer.
LiteLLM is an OpenAI compatible Proxy Server (LLM Gateway) to call 2,000+ LLM APIs using the OpenAI format Bedrock, Huggingface, VertexAI, TogetherAI, Azure OpenAI, OpenAI, etc. Get started with Opensource LiteLLM here: https://github.com/BerriAI/litellm (40,000+ Github Stars)
This is a repackaged open source software product wherein additional charges apply for hardening, security configuration, and setup support.
vLLM is a high-throughput OpenAI-compatible inference server for open-source LLMs (Python + PyTorch CPU build), bundled with Open WebUI as a browser chat front end. This Lynxroute build is secured and ready out of the box: a random 32-byte API key generated at first boot, Bearer-token auth enforced on every /v1 request, vLLM and Open WebUI bound to loopback behind Nginx TLS, a default tiny model (facebook/opt-125m) preloaded so the API and chat work immediately, UFW firewall pre-configured, and a CIS Level 1 hardened Ubuntu 24.04 LTS base.
vLLM is Apache-2.0 licensed; Open WebUI uses a source-available license (see Long Description for details).
This product has charges associated with it for the IronCloud LLM platform and support. Deploy your own secure, ChatGPT-style AI in minutes - fully private, self-hosted, and compliant with NIST 800-171, CMMC, and ITAR. IronCloud LLM runs entirely inside your AWS environment, keeping all data, prompts, and files under your control. No SaaS, no data leaks - just powerful AI you own. Built for regulated industries such as government, defense, finance, and healthcare, IronCloud LLM supports GPT-4, Claude, Gemini, AWS Bedrock, and more.
Fast inference API for open-source models like Llama, Qwen, OpenAI GPT-OSS, and more, powered by the revolutionary Cerebras Wafer-Scale Engine.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.