Listing Thumbnail

    AI Cloud Gateway - LLM Router for Amazon Bedrock

     Info
    Deployed on AWS
    LLM router for Amazon Bedrock. Each request goes to the lowest-cost Bedrock model expected to answer it well, chosen deterministically in milliseconds. Runs in your own AWS account: AWS bills Bedrock to you with no markup on tokens. Monthly budgets and optional Bedrock Guardrails. Works with the SDKs and agent frameworks you already use.

    Overview

    Open image

    AI Cloud Gateway is an LLM router for Amazon Bedrock. You send chat requests to one HTTPS endpoint, and for each request it picks the cheapest Bedrock model that is expected to answer it well, then runs it in your own AWS account.

    How it works:

    • Subscribe, sign in at dashboard.aicloudgateway.com, and create an API key.
    • Connect your AWS account with one CloudFormation stack. It creates a role that can only use Amazon Bedrock, and that we can only assume with an External ID unique to your subscription.
    • Point your client at https://api.aicloudgateway.com/v1  with your key.

    Routing: set model to auto and the router reads each request (task type, difficulty, length, tools, images) and chooses among Amazon Nova, Anthropic Claude, Meta Llama and other Bedrock models. auto:cost and auto:quality lean either way, and you can pin a single model. Add task:simple or task:complex to a prompt to move the choice, and effort:low, medium or high to make the chosen model reason harder. If a model is throttled or not enabled in your account, the request falls back to the next suitable one. Every response names the model that answered.

    How it decides: a deterministic rules engine, not another language model. For a given catalog and engine version, the same request and settings always choose the same model, in milliseconds and without an extra model call, so routing adds no token cost and can be reproduced when you audit a decision. Response headers name the model, the mode and the catalog version that made the choice.

    Controls: set a monthly cap on estimated Bedrock spend or on routed requests, for the whole subscription or for each API key; once reached, requests are refused until the next month or until you raise it. Optionally, name an Amazon Bedrock Guardrail from your own account and it is applied to every routed request.

    The API is compatible with the OpenAI API format (chat completions with streaming, tool calls, images and JSON output), so existing SDKs and agent frameworks work by changing the base URL and key, including agents on Amazon Bedrock AgentCore. AI Cloud Gateway is not affiliated with OpenAI.

    Billing: Amazon Bedrock usage is billed by AWS to your own AWS account, at your Bedrock prices. AI Cloud Gateway bills only for routing: a monthly or annual Starter, Team or Business plan, each with a monthly allowance of routed requests, then a usage charge per 1,000 routed requests beyond it at that plan's rate. Larger plans cost less per request, and refused calls are free. Each plan covers one connected AWS account. Upgrade to a larger plan during the term.

    Data handling: prompts and completions are never stored. AI Cloud Gateway keeps only usage counts (requests and tokens per model per hour) and routing decision features. Requests pass through our service in AWS us-east-1, in memory.

    Supported Bedrock regions for the connection: us-east-1, us-east-2, us-west-2.

    Highlights

    • Deterministic, cost-aware routing: each request goes to the lowest-cost Amazon Bedrock model expected to answer it well. A rules engine picks it in milliseconds, the same way every time, with no extra model call. Every response names it.
    • Runs in your AWS account: a CloudFormation role that can only use Bedrock, assumed with your External ID. AWS bills Bedrock to you at your prices, with no markup on tokens. Prompts are never stored. Optional Amazon Bedrock Guardrails.
    • Control and drop-in: monthly spend and request caps per subscription and API key, and fallback on throttling. Works with the SDKs and agent frameworks you use, including Bedrock AgentCore, by changing the base URL and key.

    Details

    Delivery method

    Deployed on AWS

    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.
    Financing for AWS Marketplace purchases

    Pricing

    AI Cloud Gateway - LLM Router for Amazon Bedrock

     Info
    Pricing is based on the duration and terms of your contract with the vendor, and additional usage. You pay upfront or in installments according to your contract terms with the vendor. This entitles you to a specified quantity of use for the contract duration. Usage-based pricing is in effect for overages or additional usage not covered in the contract. These charges are applied on top of the contract price. If you choose not to renew or replace your contract before the contract end date, access to your entitlements will expire.
    Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator  to estimate your infrastructure costs.

    1-month contract (3)

     Info
    Dimension
    Description
    Cost/month
    Starter plan
    One AWS account, email support, 50,000 routed requests a month.
    $19.00
    Team plan
    One AWS account, faster support, 1,000,000 routed requests a month.
    $139.00
    Business plan
    One AWS account, priority support, 5,000,000 routed requests a month.
    $649.00

    Additional usage costs (3)

     Info

    The following dimensions are not included in the contract terms, which will be charged based on your usage.

    Dimension
    Description
    Cost/unit
    Starter requests (1k)
    Starter: routed requests beyond 50,000 a month, per 1,000.
    $0.30
    Team requests (1k)
    Team: routed requests beyond 1,000,000 a month, per 1,000.
    $0.25
    Business requests (1k)
    Business: routed requests beyond 5,000,000 a month, per 1,000.
    $0.22

    AI Insights

     Info

    Dimensions summary

    Pricing splits into three plans: Starter, Team, and Business. Each plan is a subscription that includes a monthly allowance of routed requests. Starter includes 50,000 a month, Team includes 1,000,000, and Business includes 5,000,000. Each plan also differs in the support level you receive. Once you pass your plan's allowance, you pay per 1,000 additional routed requests through the matching overage dimension. So you pick a plan for your expected volume, then overage charges cover traffic beyond that allowance. You pay only for routing; AWS bills Bedrock tokens to you directly.

    Top-of-mind questions for buyers

    A routed request is one chat completion call sent through the gateway to a Bedrock model. The router picks the model and makes the call. Refused requests, such as those blocked by a budget cap, are not billed. You pay only for routing, not for the Bedrock tokens.
    Once you cross your plan's monthly allowance, you pay per 1,000 extra routed requests through the matching overage dimension. Only requests beyond the included amount are billed as overage. The included allowance stays part of your plan fee. Overage charges apply automatically, without a plan change.
    No. Inference runs in your own AWS account, and AWS bills Bedrock token usage to you directly at your own prices. This listing bills only for routed requests. Token charges and routing charges are separate; the routing fee does not include any token markup.
    aicloudgateway.com
    Helpful?

    Vendor refund policy

    Charges are your plan's fee plus routed requests beyond its monthly allowance, billed through AWS Marketplace. Amazon Bedrock usage is billed by AWS to your own account.

    Cancel through AWS Customer Service within 48 hours of purchase for a full refund, processed by AWS. After that the term runs to its end; turn off auto-renewal to stop billing.

    To request a refund or raise any other billing question, email contact@aicloudgateway.com  with your AWS account ID. We reply within one business day.

    How can we make this page better?

    Tell us how we can improve this page, or report an issue with this product.
    Tell us how we can improve this page, or report an issue with this product.

    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.

    Usage information

     Info

    Delivery details

    API-Based Agents & Tools

    API-Based Agents and Tools integrate through standard web protocols. Your applications can make API calls to access agent capabilities and receive responses.

    Additional details

    Usage instructions

    API

    Support

    Vendor support

    Email contact@aicloudgateway.com  or use the contact form at https://aicloudgateway.com/contact/  (no account needed). We reply within one business day, Monday to Friday.

    Include your AWS account ID. For a question about a specific call, include the x-request-id response header. Prompts and completions are never stored, so we cannot see your requests; the request id lets us trace the routing decision.

    Your subscriptions, API keys, AWS account connection and usage are at https://dashboard.aicloudgateway.com 

    Billing, cancellation and refunds run through AWS Marketplace. Amazon Bedrock usage is billed by AWS to your own AWS account. What we refund is set out in the refund policy on this listing.

    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

    Customer reviews

    Ratings and reviews

     Info
    0 ratings
    5 star
    4 star
    3 star
    2 star
    1 star
    0%
    0%
    0%
    0%
    0%
    0 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.