Genesis

Genesis is one of the three MatrixOne Intelligence platforms, alongside AI Studio and MatrixOne. It serves as the model gateway, providing governed model-calling services externally, including keys, usage, a model marketplace, and online debugging. Its internal code name is TaaS (Token as a Service). Bills and logs still use the name Genesis; both names refer to the same product.

Relationship with AI Studio and MatrixOne

Genesis provides direct access to models. After obtaining a Genesis API Key, you can call models directly from any application or script through an OpenAI-compatible protocol. Usage is charged by Token or the corresponding measurement unit to a separate Genesis Credit ledger.

AI Studio features such as agent conversations, workflows, and knowledge-base Q&A also call models internally, but use the “AI services” ledger, which has unified pricing and hides model details. Even if the Genesis model list is reused internally, this usage does not count toward the usage shown to Genesis users. MatrixOne is the data, storage, and compute foundation and does not host the Genesis key product.

Genesis (keys · model marketplace · Credit)
    │ OpenAI-compatible calls
    ├─► Your own applications / scripts
    └─► Internal platform consumption (as one of the default model sources)
              ↓
        AI Studio (internal agent/workflow calls, billed to the "AI services" ledger)

For the complete boundary among the three ledgers—AI services, Genesis, and bring-your-own model access—see Credit and billable items. To determine which ledger a call belongs to, look only at its entry point: intelligence used through the UI → AI services; models called with your own API Key → Genesis.

What the console contains

Genesis has a separate sign-in entry and console from AI Studio. Its sidebar has eight fixed entries:

Page

Purpose

Debug

Online Playground for testing chat (text/multimodal), native Messages/Responses APIs, embeddings, reranking, OCR, text-to-image, text-to-video, TTS, ASR, and other capabilities

Dashboard

Operational overview across all keys for the current user: QPM, active connections, model availability, queue wait; today’s request count, Token consumption, cost, and success rate; request volume and model health over the last 24 hours

Keys

Create, edit, disable, and delete API Keys; configure available model scope, quotas (total/daily), rate limits (RPM/TPM/concurrency), and expiration. See Manage API Keys

Usage

Usage and cost statistics by time range, provider, model, and key; Token consumption trends, model cost share, and grouped details by model/key

Logs

Detailed records for each API call: time, key, provider, model, Tokens, cost, status, duration, and source; supports viewing individual request details

Notifications

Notifications for usage, keys, calls, security, service, and other abnormal events; configurable in-app, email, and Webhook channels (DingTalk/WeCom/Lark/Slack/Discord/Teams/custom)

Models

Model marketplace: filter by type, vendor, context length, and health; view capabilities, pricing, and call examples; jump directly to Debug to test a model

Usage Guide

Integration guide covering the steps to obtain a key, Base URL, authentication, parameters and response examples for each API, and FAQs. See Integrate with the Genesis API

Obtain an API Key

The primary issuance path uses service accounts in the account system (UC). Create or select a service account under Credential Management and enable Genesis access to issue a Genesis API Key in the sk-gsa-v1-... format. The local issuance entry on the Keys page may be disabled according to platform policy; existing local keys can still be viewed and cleaned up there after it is disabled.

For the complete workflow, key controls, and lifecycle operations, see Manage API Keys. For authentication, supported endpoints, and FAQs when calling the API, see Integrate with the Genesis API.

Billing

Genesis is priced by model. Token-based APIs distinguish input, output, cache creation, and cache hits, all charged to Genesis Credit. See the model details on the console’s Models page for specific unit prices. For the billing boundary with other billable items, see Credit and billable items.

Documentation tree