> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tulipai.app/llms.txt
> Use this file to discover all available pages before exploring further.

# Provider marketplace

> How independent operators compete to serve one canonical model credit.

In version 1, any operator can register a provider offer for an active Tulip model by connecting an on-chain identity and tariff to a privately stored Hugging Face Inference Endpoint credential.

<Info>
  Hugging Face is v1's first endpoint vendor. The marketplace is designed around provider commitments and tariffs so later Gateway adapters can support additional inference vendors and GPU networks.
</Info>

## Providers can follow demand

Providers do not need to bootstrap every model speculatively. They can evaluate trading activity and market traction, decide which unserved demand is worth the cost of deployment, and register an offer when they are ready to serve it.

Once the first offer becomes active, the existing model credit becomes the inference payment asset. Additional providers can enter later and compete through independent tariffs and service quality.

## Provider offer

An on-chain offer stores:

* Model ID.
* Operator and payout addresses.
* Endpoint manifest commitment.
* Capability commitment.
* Metadata commitment.
* Current and pending tariff versions.
* Creation time and status.

The endpoint URL and access token are not exposed through Tulip's public APIs. The Gateway stores them privately and encrypts credentials at rest when PostgreSQL persistence is configured.

## Tariff

Each provider independently sets:

| Charge  | Calculation                                              |
| ------- | -------------------------------------------------------- |
| Input   | `input tokens × input credits per million / 1,000,000`   |
| Output  | `output tokens × output credits per million / 1,000,000` |
| Minimum | Lower bound on the usage charge.                         |
| Wake    | Added when the endpoint must wake from a sleeping state. |

Token-based calculations round up. The minimum applies before the wake fee.

Providers can propose a new tariff at any time, but activation is delayed. Historical tariff versions remain available so a receipt can settle against the price that was effective when the request began.

## Routing

For automatic selection, the Gateway evaluates active verified offers and chooses the lowest maximum credit charge. Ties are broken deterministically by offer ID.

Before routing, the Gateway checks:

* On-chain offer status and commitments.
* Verification freshness.
* Endpoint health and wake state.
* Model context length.
* Request capability requirements.
* Permit balance and optional offer pinning.

Automatic routing may fail over once before model output begins. Explicit offers, offer-pinned permits, and supplied offer-specific quotes do not silently reroute.

## Offer statuses

Providers may activate, pause, or retire their offers. Protocol governance may quarantine an offer. A quarantined offer returns to paused status only after governance clears it.