Vibgrate Relay Documentation

Everything you need to route VG Code through Vibgrate Relay — setup, supported models and how routing works, pricing and billing, and org-level governance controls.

Overview(1)Setup(2)Models & Routing(2)Pricing & Billing(3)Governance(1)Reference(2)

Overview

What Is Vibgrate Relay?

The hosted model router for VG Code — one account and endpoint, curated hosted models, exact per-token metering, and org-level governance.

Setup

Getting Started with Vibgrate Relay

Using Relay from VG Code

The full --provider precedence order, environment variables, and how Relay interacts with local Code Modes.

Models & Routing

Supported Models & the Relay Catalog

How the curated model catalog is built, kept current, and priced — and where Workers AI serves a model directly.

Model Selection & Routing

Relay never auto-switches your model mid-session — you pick it once, for the whole run.

Pricing & Billing

Pricing & Credits

Prepaid workspace credit, published top-up fees, and how reserve → serve → settle billing works.

BYOK: Bring Your Own Provider Key

Route through Relay for governance and audit with your own provider key — free below a generous monthly floor.

Notifications & Billing Email

What Relay tells you, when, and through which channel — balance, rate changes, and monthly statements.

Governance

Governance: Allowlists, Spend Caps & Residency

Org-level controls for which models are approved, how much a workspace can spend, and where inference runs.

Reference

Relay API Reference

The OpenAI-compatible /relay/v1 endpoints, authentication, and request/response shapes.

Troubleshooting & FAQ

Common Relay errors, what they mean, and how to resolve them.

11 articles.