QBC network · developers

Build on the network: what exists and what is planned.

This page lists every developer surface with its true status. If something is marked planned, it cannot be called yet.

Today the network records and finalises commitments for one reference AI. Serving other models is the roadmap.

APIs and SDKs.

Layer 1 node RPC

live

Substrate JSON-RPC for blocks, extrinsics and storage.

Limitation: Served by nodes SusyLabs operates.

QVM JSON-RPC (chain ID 3303)

live

EVM-compatible JSON-RPC for Solidity contracts and standard tooling.

Limitation: Single sequencer; single trusted anchor submitter.

OpenAI-compatible endpoint

live

Exists on the reference implementation (AethersMind).

Limitation: One process; confirmed on our own machines only. It serves the reference AI, not other models.

Python SDK

built

A Python SDK exists and is being rebuilt.

Limitation: Out of date, and covers the reference AI only. No receipt or verify calls.

Receipt API

planned

A hosted API that issues and checks receipts, each with an evidence-level tag.

Limitation: Not built. Milestones M1 and M3.

TypeScript SDK

planned

A TypeScript SDK alongside the rebuilt Python SDK.

Limitation: Not built. Milestone M3.

Verifier command-line tool

planned

Checks a receipt against chain state.

Limitation: Not built. Milestone M1.

Open model identity registry

planned

Any account registers a model namespace with a deposit.

Limitation: Today only validators may submit, and the chain holds one lineage. Milestone M2.

What you can do today.

Deploy and call contracts

live

The QVM accepts standard EVM tooling. Point your tools at the JSON-RPC endpoint below and use chain ID 3303.

Limitation: Single sequencer; single trusted anchor submitter.

Read the chain

live

Blocks, transactions, the model identity registry and state commitments are public.

Limitation: Served by nodes SusyLabs operates.

QVM JSON-RPC
https://qvm.qbc.network/jsonrpc
QVM chain ID
3303
Chain summary (read only)
https://api.qbc.network/chain/info

The OpenAI-compatible endpoint belongs to the reference implementation and is confirmed on our own machines only, so no public address is listed for it here.

What you cannot do yet.

  • Register your own model. Only validators may submit to the registry, and it holds one lineage.
  • Issue or check a receipt for a model call. No receipt type, receipt API or verifier tool exists.
  • Use the network with another provider's model. No external model is registered or supported.

No partnership or integration with any AI provider exists. The systems named here are examples of what the network is designed to serve, nothing more.

Documentation.

All docs →Whitepaper →QVM specification →Confidential transfers →Network economics →Run a node →Deployed contracts →Security and disclosure →GitHub ↗

Go further.

Planned: integrating other models and providers →Roadmap: milestones M1 to M6 →Evidence levels →