Documentation

LLM Hangar deploys open large language models onto GPU infrastructure in your own cloud account and gives you a private, key-authenticated, OpenAI-compatible endpoint. The infrastructure belongs to you: it runs in your account, under your provider agreement, and your provider bills you for it directly.

Quickstart

  1. Create an account and start the 7-day free trial. No credit card is needed.
  2. Connect a cloud account: AWS, Nebius, RunPod, or Verda (early access).
  3. Deploy a model from the catalog. Pick a GPU shape and region, review the cost estimate, confirm.
  4. When the deployment is ready, open its Connect tab for the endpoint URL and API key.
  5. Point any OpenAI-compatible client at it.

Guides

Connect your cloud account

AWS via a CloudFormation cross-account role, Nebius via a project-scoped service account, RunPod via an API key, Verda via client credentials. What each needs and what we verify.

Deploy a model

The deployment flow from catalog to running endpoint: shapes, regions, cost estimates, and what happens during a boot.

Use your endpoint

Working snippets for Python, curl, JavaScript, LangChain, the Vercel AI SDK, editors, and automation tools.

Budget caps and auto-destroy

Hard spend limits, self-destruct timers, wake/sleep schedules, and verified teardown.

Stop and start

States including stop_failed, the deploy-time staging choice and its cost, wake estimates, what a stopped endpoint answers, and the API.

Schedules

Weekly windows and time zones, pre-warm before the window opens, holds and wakes, webhook deliveries and signature verification.

Spot instances

Eviction per provider, automatic recovery from staged weights, the fallback and wait policies, and reacquiring spot.

Keys, sign-in and audit capture

Named keys with limits, JWT validation against your identity provider, and audit capture to a bucket you own with chains you can verify.

Connect Verda

Client credentials for a Verda project, what the adapter does with instances, volumes and startup scripts, and what the first live run measured. Early access.

Guide: deploy an LLM on AWS without the CLI

Which EC2 instance fits which model, prices by region with their sources, the quota and capacity blockers, and the four steps.

Security: access, keys and data

Where prompts go, how access to your cloud account is scoped, credential storage, the audit log, verified teardown.