platform.softquantus.com is the web console for SoftQuantus compute. It combines three things:
  • The console — where your team manages workspaces, API keys, billing, quantum instances, licenses and trust services.
  • One prepaid wallet — credits you top up once and spend on any backend.
  • The Execution Plane — the REST contract (/api/v1/platform/*) that authorizes, reserves credit for, meters and settles every workload you run: quantum circuits on simulators or real QPUs (via QCOS), and on-demand GPU/CPU jobs.
You activate compute in the console; you run workloads from your own code with the REST API, the qcoscloud Python SDK or the CLI — authenticated by the same sq-live- API key.

​
The console at a glance

​
The typical first-run path

  1. Create an account with email + password or Google.
  2. Create an API key with the scopes you need.
  3. Run your first workload — a Bell circuit on the free simulator costs nothing.
  4. Top up credits to run paid backends, and review backends and pricing.
  5. Organize access with quantum instances and your team.
  6. Verify any settled run with its Evidence Bundle.

​
Guides

Account & sign-in

Sign-up, email verification, Google sign-in, password reset, invitations.

Workspaces & team

Workspaces, the four roles, invitations and switching.

API keys

Scoped sq-live- keys for the REST API, SDK, CLI and MCP.

Billing & credits

Prepaid wallet, top-ups, spend limits and invoices.

Quantum instances

Free, prepaid and pay-as-you-go plans, target allowlists, access groups.

Run workloads

Quickstart: a Bell circuit over REST, plus GPU and CPU jobs.

Backends & pricing

Simulators, QPUs, GPU/CPU tiers and how cost safety works.

REST API

The live endpoint reference for /api/v1/platform/*.

Python SDK

The qcoscloud client: credits, instances, circuits, GPU/CPU jobs.

CLI

The qcoscloud command line for management and GPU/CPU jobs.

Evidence Bundles

Hash-pinned execution evidence and what result_verified means.

Errors & refusal reasons

Every machine reason, its HTTP status and how to fix it.

​
Cost safety, in one paragraph

Every paid run reserves its worst-case cost before anything is submitted, settles the actual metered cost when it finishes, and releases the unused reservation. Per-minute GPU/CPU jobs require a runtime bound and are stopped when their reservation is exhausted, and a backend that is not available fails closed with no money held. Details in Backends & pricing.