Sold by: SpendProxy
Deployed on AWS
A self-hosted proxy that gives engineering teams cost visibility, attribution, and spend controls over their LLM API usage across providers.
Overview
SpendProxy is a self-hosted proxy for teams running production workloads against large language models. It sits inline between your applications and the LLM providers, giving you cost visibility, per-team and per-project attribution, and spend controls behind a single endpoint.
It works across OpenAI, Anthropic, Google, and Amazon Bedrock. Your requests are forwarded to whichever of those providers you configure; SpendProxy keeps the cost, usage, and attribution data it derives in storage you control, in your own AWS account. Provider API keys and that operational data stay in your infrastructure - your prompts and responses still travel to the LLM providers you call, as they would without a proxy.
SpendProxy is designed to be safe on the request path: a control-plane problem (including an entitlement-check failure) never blocks your production calls - it fails open.
Capabilities:
- Cost tracking and historical reporting across providers
- Per-team / per-project / per-key cost attribution
- Budget guardrails, model routing, caching, and retry-storm suppression (opt-in)
- Security gating, rate limits, and an audit trail for LLM access
- Accuracy benchmarking and what-if cost simulation
Highlights
- One endpoint across OpenAI, Anthropic, Google, and Amazon Bedrock - cost visibility and per-team attribution.
- Self-hosted in your own AWS account/VPC - provider keys and your cost and usage data stay in your infrastructure.
- Fail-open on the request path - a control-plane or entitlement-check failure never blocks your production traffic.
Details
Sold by
Delivery method
Supported services
Delivery option
Amazon ECS (container image)
Latest version
Operating system
Linux
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 the duration and terms of your contract with the vendor. This entitles you to a specified quantity of use for the contract duration. If you choose not to renew or replace your contract before it ends, access to these entitlements will expire.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/month |
|---|---|---|
SpendProxy - monthly subscription (1 deployment) | Monthly subscription to SpendProxy, the LLM cost-control and governance proxy, for one self-hosted deployment. Includes the dashboard, per-team and per-project cost attribution, and spend controls. | $2,999.00 |
Dimensions summary
This listing offers a single pricing option: a monthly subscription that covers one self-hosted deployment. You pay a flat recurring fee per deployment, so pricing scales by the number of deployments you run rather than by usage or request volume. The subscription includes the dashboard, per-team and per-project cost attribution, and spend controls. There are no separate tiers or add-ons to choose from. To cover more than one deployment, you add subscriptions.
Top-of-mind questions for buyers
One deployment is a single installation of the SpendProxy container running in your own infrastructure. It sits in front of your AI APIs as a reverse proxy. The subscription covers that one running instance. To cover more than one deployment, you add more subscriptions.
No. You pay a flat recurring fee per deployment. The subscription does not meter the number of requests passing through the proxy or your underlying AI provider spend. Your software cost stays the same whether traffic rises or falls. Only adding deployments changes the subscription cost.
The subscription includes the dashboard, per-team and per-project cost attribution, and spend controls. The five optimization engines run with the deployment. There are no separate tiers or add-ons to buy. Each engine runs in off, monitor, or autopilot mode, which you configure yourself.
spendproxy.com+1
Helpful?
Vendor refund policy
Contract fees are generally non-refundable. AWS's 48-hour cancellation window for public contracts applies, and private-offer terms govern where applicable. To stop future charges, disable renewal before the renewal deadline. For refund requests, contact support@spendproxy.com .
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
Amazon ECS (container image)
Supported services: Learn more
- Amazon ECS
Container image
Containers are lightweight, portable execution environments that wrap server application software in a filesystem that includes everything it needs to run. Container applications run on supported container runtimes and orchestration services, such as Amazon Elastic Container Service (Amazon ECS) or Amazon Elastic Kubernetes Service (Amazon EKS). Both eliminate the need for you to install and operate your own container orchestration software by managing and scheduling containers on a scalable cluster of virtual machines.
Version release notes
Initial release of SpendProxy - a self-hosted proxy for LLM cost control and governance. Records per-feature cost that reconciles to the provider bill, with prompt caching, model routing, and budget controls. Supports Anthropic, Amazon Bedrock, OpenAI, and Google.
Additional details
Usage instructions
SpendProxy is a self-hosted proxy between your application and its LLM providers (Anthropic, Amazon Bedrock, OpenAI, Google). It records per-feature cost that reconciles to the provider bill, and adds prompt caching, model routing and budget caps. It runs inside your own VPC; traffic and data never leave your account. One container listens on port 4100 and serves a cost dashboard at /dashboard.
-
PULL THE IMAGE (after subscribing) 709825985650.dkr.ecr.us-east-1.amazonaws.com/spendproxy/spendproxy:0.1.0 Subscribing grants your AWS account pull access. An ECS task execution role with the standard AmazonECSTaskExecutionRolePolicy already has the required ECR pull permissions.
-
REQUIRED ENVIRONMENT VARIABLES CE_ENTITLEMENT_MODE=aws-marketplace (required; no separate license key is needed in this mode) CE_ENTITLEMENT_REGION=us-east-1 CE_MARKETPLACE_PRODUCT_SKU=prod-6bhzbkuyw5jso CE_PROXY_PORT=4100
-
TASK ROLE IAM license-manager:CheckoutLicense validates your subscription with a PROVISIONAL check that consumes no units. The check is fail-open and gates only the dashboard. Add provider permissions for the routes you use, for example bedrock:InvokeModel and bedrock:InvokeModelWithResponseStream for the /bedrock route. The Anthropic, OpenAI and Google routes forward your own API key and need no extra IAM.
-
DASHBOARD AUTHENTICATION (choose one) CE_DASHBOARD_PASSWORD=<secret> to password-protect the dashboard, or CE_DASHBOARD_AUTH=alb-oidc with CE_DASHBOARD_ALB_ARN=<alb-arn> to verify an OIDC-authenticating ALB (no password stored in the container).
-
OPTIONAL STATE PERSISTENCE Mount a volume at CE_DATA_DIR, or set LITESTREAM_REPLICA_URL=s3://<bucket>/<prefix> to replicate the SQLite cost database to S3 (the task role then needs read/write on that bucket).
HEALTH CHECK: HTTP GET /health on port 4100.
ROUTING YOUR APPLICATION Anthropic: set the SDK base URL to http://<spendproxy-host>:4100/anthropic Bedrock: set the runtime endpoint to http://<spendproxy-host>:4100/bedrock OpenAI: http://<spendproxy-host>:4100/v1 Google: http://<spendproxy-host>:4100/google
Support
Vendor support
Email support@spendproxy.com for setup, deployment, and entitlement/billing questions. Response target: one business day.
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)
MAI Gateway unifies enterprise LLM governance with budget control, intelligent routing, audit trails, and security compliance for AI workloads.
LLM Provost for Compliant Organizations like Medical or Market trading. A mandatory policy enforcement and observability boundary that sits in front of MCP servers and upstream model/tool APIs.
Professional service to build AI model invocation dashboards on AWS, visualizing usage, latency, errors, token consumption, and cost across Amazon Bedrock and related services.
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.