WolfPackAI

One gateway for every model your teams use.

Alpha · pilots available

An AI gateway for model routing, workflows, and tracing. Run it on your servers or private cloud. Local models stay on your hardware; only requests you choose to route to commercial model providers leave your environment.

Book a discovery call See the stack
Runs on
Your servers or private cloud
What leaves
Only the requests you choose to route to commercial model providers. Local models stay on your hardware.
Works today
Core orchestration, basic authentication, and model routing.
Roadmap
Role-based access control, richer monitoring, and Kubernetes deployment.
Access
Alpha; pilots available through a discovery call

Choose the components for your deployment.

WolfPackAI brings these tools together on your infrastructure. The composition depends on your models, workflows, and operating requirements.

OpenWebUI

A clean portal for chat and day-to-day model use across your teams.

LiteLLM

One gateway to open and closed model providers—without rewriting every integration.

n8n

Workflow automation for connecting models to your systems. Groundwork is built; full n8n integration is still in progress.

Ollama

Local LLMs on your GPUs when data residency and latency demand it.

Observability

OpenTelemetry traces, metrics, and logs in the Aspire dashboard so you can see what each service is doing.

.NET Aspire + Docker

Composable microservices you can deploy and evolve with your platform team.

Start with a model routing pilot.

  • Routing across local and commercial model providers through one gateway
  • Workflow automation and webhooks through n8n, with full integration still in progress
  • Deployment on existing hardware, new infrastructure, cloud, or hybrid
  • Local processing or commercial API requests, according to the routes you configure
  • A pilot to assess the work needed before production

Current status

WolfPackAI is in alpha. Core orchestration, basic authentication, and model routing work today and are ready for pilots.

Role-based access control, richer monitoring, and Kubernetes deployment are roadmap work. Do not plan a deployment that depends on those features being available today.

From discovery to a working gateway.

  1. Run a pilot

    Discovery, environment choice, platform install, first model routes live, and a clear read on what your teams will actually use.

  2. Review production requirements

    Identify the access controls, monitoring, workflow changes, and operator training your deployment needs. Review the gaps with your security and compliance teams before expanding.

Where it fits: law firms, banks and credit unions, and healthcare AI governance. It can also host the model side of AI agents and private RAG.

WolfPackAI sits beside your existing systems. An alpha pilot is not a complete policy enforcement system or a compliance certification. People remain responsible for consequential actions.

Put the gateway on your infrastructure.

Schedule a discovery session. We will map WolfPackAI to your hardware, identity, and model strategy.

Book a discovery call